Chap. 3.1 AAO and Mask
ChChaapp.. 33..11..11 AAAAOO
As introduction of AAO at Chap. 1.3, we should take notices on some factors in order to prepare a standard which conformed to our requirements.
Those key points were classed as and the process details of anodization.
First, the factors of aluminous intrinsic quality were purity and crystallization enough or not. The purity was depended on the Al sheet or ingot for evaporation. For Al sheet, we employed the high purity as well as we can. However, it was still not good enough to form AAO due to problems about crystalline of Al. We should enhance the crystalline quality by annealing. The factors in annealing aspect were such as atmosphere, length of time, temperature. As lots of reports, we knew intuitively that higher temperature, under inert gas atmosphere, and longer time would improve batter crystalline quality. Exactly, the re-crystallization of Al caused of the suitable annealing. We considered the melting point of Al, environment friendliness, and process integration, so we employed suitable parameters including annealing temperature:
300~500 , time: 2℃ -8hrs, in Ar atmosphere.
Second, we focused on the parameters of anodic process including different kinds of electrolyte, reaction temperature, applied polarization voltage, reaction time, widening time, removed time, supersonic shake time, and concentration about above every kinds of chemical solutions. As
previous studies, we employed suitable parameters, list as following: Al sheet or Al/TiN/Si-sub, applied voltage: 30, 40, 50V, space of electros: 7 units, anodic reaction T: 12℃. The reaction time and widening time were depended on geometry scale we required; for example, we observed the porous length of 1μm by total reaction time of 6 min in our two-step anodization and the diameter of 80nm by widening time of 40 mins in our recipes. However, AAO was fabricated by improved recipes. For example, the supersonic shake procedure removed the fracture on the AAO surface;
if conventional two-step anodization replaced by multiple-step anodization, we avoided supersonic shake procedure and obtained satisfied products.
Following would show some images of measurement and discuss with those data.
Fig. 3.1-1 depicts the SEM top-view of AAO array morphology formed by 4Ni Al sheet with different annealing temperature. We observed the porous array along a direction in all images and the different phenomenon by the same widening recipe. It appears the intrinsic quality of Al is the major cause, and we suggest the distance is formed by the machine stretch. It can be solved by chemical polish or machine polish.
Fig. 3.1-2 shows the relation of widened time and AAO morphology.
We can confirm that the widening time with porous diameter in proportion.
The over widening time was applied; the AAO became brush-liked AAO (Fig. 3.1-3) which was composed by fracture pores on the AAO surface.
The structure has been studied34.
Fig. 3.1-4 and Fig. 3.1-5 reveal AAO structures fabricated by multiple step anodization. We observed the more fracture pores on the AAO surface by more step anodization. We observed further the ideal AAO pores array under the fracture pores film in Fig. 3.1-5(a), (c).Fig. 3.1-6 reveals the result of AAO array after supersonic shake procedure. The result equaled to AAO after widening and supersonic shake procedure. We obtained the satisfied AAO array, and the MASK based on the structure.
Fig. 3.1-7 indicates the AAO structures fabricated with the same parameters expect electrolytes. (a) was fabricated by 0.3M oxalic and (b) was fabricated by 0.1M H3PO4 , and both without widening treatment. We demonstrate the porous diameter depends on the kind of electrolytes.
(a) (b)
(e) (f)
(c)
(g) (h)
(d)
(a) (b)
(c) (d)
(e)
Fig. 3.1-2: 4N Al sheet; annealing 400 ℃ x 2hrs; applied voltage 50V; space of electros: 7; reaction T: 12℃
Different widen time
NO a b c d e Widen
time 0 20 40 60 100
(a)
(
Fig. 3.1-3: The brush-liked AAO was fabricated by over widening (widened time
Fig. 3.1-4: They were fabricated by conventional two-step anodization. (a) SEM image only by fist step anodize (6, 1), (b) Cross section view. (c) SEM image by two-step anodize (6, 1, 3, 1), (d) Cross section view. (4N Al sheet; annealing 400 x 8hrs; applied voltage 40V; space of ℃ electros: 7; reaction T: 28℃)
(a) (b)
(c) (d)
Fig. 3.1-5: They were fabricated by multiple step anodization which was more steps then two-step anodize. (a) SEM image by three step anodize (6, 1, 3, 1, 3, 1), (b) Cross section view. (c) SEM image by four step anodize (6, 1, 3, 1, 3, 1, 3, 1), (d) Cross section view. (4N Al sheet; annealing 400 x 8hrs; applied ℃ voltage 40V; space of electros: 7; reaction T: 28℃)
(a) (b)
(c) (d)
Fig. 3.1-6: It was cleaned the residuum on the surface by supersonic shock after multiple step anodize. (a) High magnification (100K) SEM image and (b) low magnification (50K) SEM image revealed the uniform pore array. (4N Al sheet; annealing 400 x 8hrs; applied voltage 40V; space of ℃ electros: 7;
reaction T: 28℃) (a)
(b)
Fig. 3.1-7: They were fabricated with the same parameters expect electrolytes. SEM images of the same magnification (100K), (a) was fabricated by 0.3M oxalic and (b) was fabricated by 0.1M H3PO4. (4N Al sheet; annealing 400 x ℃ 2hrs; applied voltage 40V; space of electros: 7; reaction T: 28℃)
(a)
(b)
ChChaapp.. 33..11..22 MaMasskk
Based on the AAO structure, we obtained the MASK by removed the barrier (bottom) layer. The key point was on the recipe of removing bottom and widening.
Fig. 3.1-8 shows the SEM images of MASK, including top-view, back-view, and cross-section view. When removing time was increased, we observed that the thickness of bottom was thinner and the porous wall was thinner simultaneously.
Large area (more than 1cm2) mask would be fabricated by the suitable time combinations with widening and different removing bottom time (Fig. 3.1-9 (c)). Otherwise, we observed partial porous bottom were removed and partial ones were remained (Fig. 3.1-9 (a), (b)), although the (a) and (b) carried out by more reaction time than (a). Based on the isotropic chemical etching, it seems be controlled by the reaction ratio due to contact surface. And the uniformity of the top surface is a problem, it maybe roughness caused of fringe-liked residues of AAO.
Fig. 3.1-8: AAO template (mask) was fabricated by removing bottom layer after removed superfluous Al film with AAO. (a) as-prepared AAO top view of SEM image. (b) The cross section of AAO after removed superfluous Al
(a) (b)
(c) (d)
(e) (f)
Fig. 3.1-9: SEM images about back view of AAO with different reaction time combinations which were different widening and different removing bottom time. (a) total time 100 (70+30) mins, (b) 80 (40+40) mins, (c) 60 (0+60)mins. (widening time, removing time)
( ( (b)
Chap. 3.2 ZnO with AAO
Based on the recipe of fabrication AAO on Al sheet, we fabricated on Al/TiN/Si-sub in order to integrate into VLS and increase the applications.
Following, the section lists results and discussions including SEM, PL, XRD, TEM, FE measurements.
ChChaapp.. 33..22..11 SESEMM AAnnaallyysisiss
Fig. 3.2-1 shows the relation of widened time and AAO morphology.
We can confirm that the widening time with porous diameter in proportion according to fabricated once on Al sheet. So we control the widening time to achieve the required aspect ration. However, because of the Al film formed by 4N Al ingot, we observed residues on the surface and effect on porous direction. We suggest the residues formed by impurities precipitated after annealing treatment.
Fig. 3.2-2 reveals the top view of AAO and many residues remain on surface because of the quality Al film was no good enough. And (c) is the high magnification (150K) image indicates porous diameter was about 60-85nm, thickness of porous well was about 50 nm and thickness of bottom was about 50 nm. The thickness of porous well equaled to thickness of bottom. The details of structure was AAO/Al/TiN/Si-sub, the
Fig. 3.2-3Fig. 3.2-3 reveals the bottom layer becomes thinner after over reaction treatment. Relatively, the pores become fracture. We suggest that Si and Ti would tend to from silicide after 400℃annealing, maybe the Al and Ti formed alloy, too. So, we observed a layer existed under the AAO after over suitable second anodic time.
Fig. 3.2-4 reveals clearly the cross-section view of pores full of ZnO successfully, (a): AAO was fabricated by conventional two-step anodization (3, 12) and widened in 60 mins. Diameter of pores was about 80nm, (b): It was removed the ZnO residues on AAO surface after pores full of ZnO by hydrothermal method. Then we removed the AAO by chemical etching in HNO3 solution. It is showed at Fig. 3.2-5 (a) is the top view of SEM image reveals island-liked ZnO NWs crowd. (b) is the image of cross section view indicates the collapse of ZnO NWs clearly. We observed some broke NWs which stood on the TiN layer vertically, and the other ones which were intact structure were merge together.
Fig. 3.2-6 reveals SEM images of that AAO was fabricated by conventional two-step anodization. The special treatment was supersonic shake to remove residues on surface after widening. We observed the better structure then Fig. 3.2-2, but the porous length was shorter as a result to sacrifice a litter AAO film when the supersonic shake treatment.
As the Fig. 3.2-7, it reveals the different surface morphology of Al film which was evaporated by 4N and 5N Al ingot respectively. We observed lots of impurities on the AAO surface formed by 4N Al source, and another was clean and uniform. Further, the high purity of Al film would provide higher uniform, ordered, vertical pores for applications.
If the reaction temperature of hydrothermal method is too low, we observed those ions could not successfully migrate into pores. Fig. 3.2-8 indicates clearly that the Zno crystallized at top of pores when the reaction temperature was 80 . ℃ The causes maybe include the morphology of AAO and the geometry of pores except the reaction temperature; we suggest the reaction temperature is the principal cause.
Fig. 3.2-9 and Fig. 3.2-10 reveal the morphology of AAO formed by 5N Al film. At
Fig. 3.2-9, (a), (b), (c) reveal different applied voltage to anodize
dominate the growth ration of AAO and do not affect the scale of porous diameter.
Fig. 3.2-11 reveals the image which the sample after field emission measurement. (a) is the original structure, diameter: 20-30nm, length:
1800nm. And (b), (c), (d) reveal the broken area after field emission measurement.
Fig. 3.2-1: These AAO were fabricated by two-step anodization (3, 3) on SiO2/Si-sub.
(a), (b) were AAO SEM images without widening treatment, porous of diameter was small then 10nm. (c) (d) were widened in 20 mins, porous of diameter was about 30nm. And (d) (e) were widened in 30 mins, porous of
(a) (b)
(c) (d)
(e) (f)
Fig. 3.2-2: AAO was fabricated by conventional two-step anodization (3, 3) and widening time was 40mins. (a) It revealed the top view of AAO and many residues on surface. (b) cross section view image revealed the AAO thickness was about 1.2μ. (c) The high magnification (150K) image indicated porous diameter was about 60-85nm, thickness of porous well was about 50 nm and thickness of bottom was about 50 nm. The thickness of porous well equaled to thickness of bottom. (4N Al /SiO2/Si; thickness of Al:
1μm; annealing 350 x 8hrs; applied voltage: 40V; ℃ electrolyte: 0.3M oxalic; space of electros: 7; reaction T: 12℃)
(a) (b)
(c)
(a)
(b)
Fig. 3.2-3: AAO was fabricated by conventional two-step anodization (3, 12). They were anodized over suitable time (about 3min) at second anodic treatment.
(a) it was widened in 50 mins. (b) it was widened in 80 mins, we observed over winding induced porous wall to break and thickness of bottom to thinner. (4N Al /TiN/Si; thickness of Al: 1μm; annealing 400 x 8hrs; ℃
(a)
(b)
Fig. 3.2-4: (a) AAO was fabricated by conventional two-step anodization (3, 12) and widened in 60 mins. Diameter of pores was about 80nm. (b) It was removed the ZnO residues on AAO surface after pores full of ZnO by hydrothermal method. The SEM image revealed the pores full of ZnO successfully. (4N Al /TiN/Si; thickness of Al: 1.3μm; annealing 400 x 2hrs; applied voltage: ℃ 40V; electrolyte: 0.3M oxalic; space of electros: 7; reaction T: 12 ; ZnO ℃ hydrothermal in 2hrs at 96℃)
(a)
(b)
Fig. 3.2-5: The AAO full of ZnO was removed AAO in order to form the exposed ZnO NWs. (a). the top view of SEM image revealed island-liked ZnO NWs crowd. (b). the image of cross section view indicated the collapse of ZnO NWs clearly. (4N Al /TiN/Si; thickness of Al: 1.3μm; annealing 400 x ℃ 2hrs; applied voltage: 40V; electrolyte: 0.3M oxalic; space of electros: 7;
(a)
(b) (c)
Fig. 3.2-6: AAO was fabricated by conventional two-step anodization. The special treatment was supersonic shake to remove residues on surface after widening. (a) Top view. (b)Cross-section view and (c) was high magnification (250K) image of the bottom area. (4N Al /TiN/Si; thickness of Al: 1μm; annealing 400 x 8hrs; app℃ lied voltage: 40V; electrolyte: 0.3M oxalic; space of electros: 7; reaction T: 12℃)
(a)
(b)
Fig. 3.2-7: AAO was fabricated by two-step anodization. The top-view SEM images were the same magnification, (a) employed 5N Al source and (b) employed 4N Al source. We could clearly observe the improvement of reduced
(a)
(b) (c)
Fig. 3.2-8: SEM images of cross section(a),top view(b)(c) Temperature not high enough to assist ZnO migrated into pores, so we observed the ZnO covered the surface and would not full into the porous.(b) 30 mins at 80℃,(c) 1hr at 80℃. The inset in (a) was the high magnification SEM image.
(b) (a)
(d) (c)
(g) (f)
Fig. 3.2-9: SEM images of cross-section view (a)(b)(c) were different applied voltage to anodize according to 30V, 40V, 50V. (d), (e), and (f) were reveled different widening time form the sample of (c), and the widening time was
(a) (b)
(c) (d)
Fig. 3.2-10: (a), (c) are the SEM images of top view, and (b), (d) are SEM images of cross-section view. (a), (b) are fabricated by convectional two-step anodization.(1, 3.5) (c), (d) are widened in 60 mins, the diameter become about 80 nm from 10 nm. (5N Al /TiN/Si; thickness of Al: 500nm;
annealing 400 x 2hrs; applied voltage℃ : 50V; electrolyte: 0.3M oxalic;
space of electros: 7; reaction T: 12℃)
(b) (a)
(c) (d)
(b)
Fig. 3.2-11: They are SEM images of cross-section view, (a) is the original structure, diameter: 20-30nm, length: 1800nm. And (b), (c), (d) reveal the broken area after field emission measurement. (5N Al /TiN/Si; thickness of Al: 1500nm;
annealing 400 x ℃ 2hrs; applied voltage: 50V; electrolyte: 0.3M oxalic;
space of electros: 7; reaction T: 12 ; widening time 20℃ )
ChChaapp.. 33..22..22 TETEMM AAnnaallyyssiiss
About the analysis of TEM, we show data at following. Moreover, we prepared the sample for TEM by FIB (focus ion beam) and supersonic shake method.
Fig. 3.2-12 reveals the high resolution image about the cross-section view of AAO full of ZnO. Fig. 3.2-13 reveals: (a) the TEM image of ZnO NW by supersonic shake treatment; (b) the SAED of ZnO NW. (c) the EDS of ZnO NW.
We observe the SAED pattern which seems the type of crystallization mixes with amorphous and poly. We suggest the crystalline of ZnO maybe a poly type, and the amorphous signal from AAO according to the Al2O3
of AAO is amorphous type. The result is expectable due to the structure of ZnO NW, prepared by supersonic shake treatment, was a ZnO wire covered with a Al2O3 film, the EDS shows the concentration of Zn is lower.
(a)
TEM images
Fig. 3.2-12: The TEM cross-section image of AAO full of ZnO prepared by FIB; ; the porous diameter is about 80nm after 40mins widening treatment; the structure is AAO/SiO2/Si-sub.
(a) (b)
(c)
Fig. 3.2-13: (a) TEM image of ZnO NW prepared by supersonic shake treatment; (b) SAED of ZnO NWs.(c) EDS of ZnO NWs.
ChChaapp.. 33..22..33 XRXRDD AAnnaallyysisiss
The crystal structure and phase of the NWs were determined by X-ray diffraction (XRD, MAC Science, MXP18, and Japan). Fig. 3.2-14 depicts the XRD patterns of Al thin film under different annealing treatment on TiN/Si-sub, it compares with (a) different annealing time and temperature, (b) RTA, respectively. The major annealing treatment was at 300~500 °C for 1~8 hrs in the Ar atmosphere, and cool down to room temperature in chamber naturally. Another one was rapid thermally annealed (RTA) at 400 or 500 °C for 3 min in the Ar atmosphere. We observe the peak of TiN disappeared due to the higher T and longer time annealing or RTA. It seems to imply that TiN was composed with Al or the quality of TiN deposited by sputter was not good. Actually, we observed interface (between Al film and TiN) adhesion after RTA treatment was inferior to ones after long time annealing. Consequently, we elected some suitable parameter for annealing which annealing was carried out at 400 ℃ for 2hrs, and replaced barrier layer of TiN with SiO2 layer.
At Fig. 3.2-15, those XRD patterns of AAO full of ZnO on SiO2/Si-sub indicate the different crystalline of sample with annealing or not. The XRD peak at around 33° is caused by the ZnO crystal, and the main peaks indexed as (002) of the wurtzite structure. The batter crystalline of ZnO would exhibit better FE properties according to
Fig. 3.2-14: It reveals XRD patterns of Al film with different annealing treatment on TiN/Si-sub., and compares with (a) different annealing time and temperature, (b) RTA.
30 35 40 45 50 55 60 0
50 100 150 200 250
300 ASS4.3WF
ASS4.2WFA X-Ray intensity
(abs-unit)
Al
2 Theta(degree)
AAO/SiO2/Si-sub
Fig. 3.2-15: It reveals XRD patterns of AAO full of ZnO on SiO2/Si-sub., and compares with (a) annealing (400℃ 2hr) and (b)without annealing.
ChChaapp.. 33..22..44 PLPL AAnnaallyysisiss
A photoluminescence analyzer (PL, Hitachi F-4500, Japan) with Xe lamp as an excitation source (320 nm) was used for optical studies at room temperature. Form the reaction equations (Chap. 2.2.2), the Zn2+
ions are sufficient in this solution, but the OH- ions are finite in the solution because hexamethylenetetramine (HMTA) slowly releases OH-. Thus, oxygen is expected to be lacking in the NWs fabricated by the hydrothermal method. This phenomenon also results in the deep-level35 and singly ionized oxygen vacancies36 luminescence at 495 and 520 nm as shown in Fig. 3.2-16.
Fig. 3.2-16: It reveals the with Xe lamp as an excitation source (320 nm) at room-tempurature.
350 400 450 500 550 600 0
350 400 450 500 550 600 0
ChChaapp.. 33..22..55 FiFieelldd EEmmisissisioonn AAnnaallyysisiss
First, we show a Fowler-Nordheim (F-N) function, which details were introduced at section ch 1.2.3, to analysis Field emission of our sample.
The field emitting enhancement factor, β, exhibits the effect of field emission by quantification. Value of β is calculated by fitting the slop of ln(J/E2)-1/E curve( name F-N plot). It illustrates at Fig. 3.2-17. We recommend the concept of operation region37 for analysis further, which illustrates at FIG- 3.2-18. If the length is increased, the resistance property of NW would dominant the FE property early.
◎ F-N function fitting
1. Fowler-Nordheim (FN) equation:
J = (Aβ2E2/Φ) exp(-BΦ3/2 /βE)
A：1.56x10-10 (AV-2eV) ; B：6.83x103(VeV-3/2μm-1)
2. The field emitting enhancement factor, β: (slop is fitted in FN plot) Slope =BΦ3/2d/β Æ β= BΦ3/2/S
◎ Operation region
1. Off region 2. Turn-on (Active) region 3. Saturation region
1 2 3 4
Fig. 3.2-17: Field emitting J-E curve and ln(J/E2)-1/E curve.
Fig. 3.2-18: It illustrates operation region of field emission.
0 1 2 3 4
Off region Turn-on region
Saturation
Following, we show some results and discuss characteristic of filed emission. Further, we utilize characteristic of AAO to control the structure of ZnO NWs and study the effect of aspect ratio.
First, the flower-liked (nanosheet) ZnO structure shows at Fig. 3.2-19;
it also exhibits characteristic of filed emission, and it’s effect is better than NWs in the same scale actually20. The major cause is the screen effect due to higher number density of NWs.
By comparing with Fig. 3.2-20 and Fig. 3.2-21, we observe the better value of β which becomes 2800 from 1600 and the higher current density which is enhanced one order. The photoenhanced field emission characteristic under 30W incandescent lamp irradiation is demonstrated the influence of the illumination on the field emission.
Fig. 3.2-19: The SEM images reveals the morphology of sample B2. It was AAO full
2 4 6 8 10
Fig. 3.2-20: It shows field emission J-E curve and F-N plot of sample B2. The table shows the detail data.
electrode-A(cm2) 0.00709
thickness(um) 500 emitting factor-β 1630
2 4 6 8 10
Fig. 3.2-21: It shows field emission J-E curve and F-N plot of sample B2 under a 30W lamp. The table shows the detail data. It indicates the photo-enhanced field emission characteristic.
electrode-A(cm2) 0.00709
thickness(um) 500
emitting factor-β 2800 842
Furthermore, we discuss the effects of annealing on field emission property. At Fig. 3.2-22, the SEM images reveals the morphology of sample C1; clearly, it indicates the length of ZnO NW is about 500nm, and Fig. 3.2-23 reveals that it exhibits better field emission property than flower-liked ZnO. The causes maybe are better crystallization, morphology and high aspect ratio.
Through by comparing with Fig. 3.2-24, we observe the better value of β which becomes 3660 from 2200. We suggest the cause is the better crystallization due to annealing treatment. By the way, we didn’t observe
Through by comparing with Fig. 3.2-24, we observe the better value of β which becomes 3660 from 2200. We suggest the cause is the better crystallization due to annealing treatment. By the way, we didn’t observe