Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
GROWTH OF AI2O3 THIN FILMS FOR PHOTOVOLTAIC APPLICATIONS
Document Type and Number:
WIPO Patent Application WO/2012/028534
Kind Code:
A1
Abstract:
A composition comprising, for 100% of its total mass, at least 97% of an aluminium precursor of formula: Al(R1)2(OR2) with R1 and R2 independently selected in the group consisting of methyl, ethyl, propyl, isopropyl and t-butyl; and : - From 200 ppb to 5 ppm of Mo (Molybdenum); - From 1000 ppb to 5 ppm of Fe (Iron); - From 200 ppb to 5 ppm of Cu (Copper); - From 200 ppb to 5 ppm of Ta (Tantalum).

Inventors:
BLASCO NICOLAS (FR)
DINGEMANS GIJS (NL)
KESSELS WILHELMUS MATHIJS MARIE (NL)
LACHAUD CHRISTOPHE (FR)
MADEC ALAIN (FR)
POTTS STEPHEN (NL)
Application Number:
PCT/EP2011/064656
Publication Date:
March 08, 2012
Filing Date:
August 25, 2011
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
AIR LIQUIDE
UNIV EINDHOVEN TECH (NL)
BLASCO NICOLAS (FR)
DINGEMANS GIJS (NL)
KESSELS WILHELMUS MATHIJS MARIE (NL)
LACHAUD CHRISTOPHE (FR)
MADEC ALAIN (FR)
POTTS STEPHEN (NL)
International Classes:
C23C16/40; C07F5/06; C23C16/455; H01L31/0216
Other References:
G. DINGEMANSR. SEGUINP. ENGELHARTM.C.M. VAN DE SANDENW.M.M. KESSELS: "silicon surface passivation by ultrathin AI203 films synthesized by thermal and plasma atomic layer deposition", PHYS. STATUS SOLIDI, vol. RRL4, no. 1-2, February 2010 (2010-02-01), pages 10 - 12, XP002619642
KOH ET AL: "Chemical vapor deposition of Al2O3 films using highly volatile single sources", THIN SOLID FILMS, ELSEVIER-SEQUOIA S.A. LAUSANNE, CH, vol. 304, no. 1-2, 1 July 1997 (1997-07-01), pages 222 - 224, XP005278719, ISSN: 0040-6090, DOI: DOI:10.1016/S0040-6090(97)00132-6
CHO WONTAE ET AL: "Atomic layer deposition of Al2O3 thin films using dimethylaluminum isopropoxide and water", JOURNAL OF VACUUM SCIENCE AND TECHNOLOGY: PART A, AVS /AIP, MELVILLE, NY., US, vol. 21, no. 4, 1 July 2003 (2003-07-01), pages 1366 - 1370, XP012006440, ISSN: 0734-2101, DOI: DOI:10.1116/1.1562184
P.VITANOV: "Chemical Deposition of AI203 thin films on Si substrates", THIN SOLID FILMS, vol. 517, 2009, pages 6327 - 6330, XP026459599
HOEX ET AL.: "Silicon surface passivation by atomic layer deposited A1203", JOURNAL OF APPLIED PHYSICS, vol. 104, 2008, pages 044903
G. DINGEMANS, R. SEGUIN, P. ENGELHART, M.C.M. VAN DE SANDEN, W.M.M. KESSELS: "silicon surface passivation by ultrathin AI203 films synthesized by thermal and plasma atomic layer deposition", PHYS. STATUS SOLIDI RRL4, 2010, pages 10 - 12, XP002619642, DOI: doi:10.1002/pssr.200903334
V.G. LITOVCHENKO, N.I. KLYUIS, A.A. EVTUKH, A.A. EFREMOV, A.V. SARIKOV ET AL.: "Solar cells prepared on multicrystalline silicon subjected to new gettering and passivation treatments", SOLAR ENERGY MATERIALS AND SOLAR CELLS, vol. 72, 2002, pages 343 - 351, XP004339782, DOI: doi:10.1016/S0927-0248(01)00182-9
ZOTH ET AL.: "Analytical and Diagnostic Techniques for Semiconductor Materials, Devices and Processes", 1999, THE ELECTROCHEMICAL SOCIETY PROCEEDINGS, article "Silicon contamination control by lifetime measurements"
A.A. ISTRATOV, C. FLINK, H. HIESLMAIR, S.A. MCHUGO, E.R. WEBER: "Diffusion, solubility and gettring of copper in silicon", MATERIALS SCIENCE AND ENGINEERING, vol. B72, 2000, pages 99 - 104, XP004192067, DOI: doi:10.1016/S0921-5107(99)00514-0
AGOSTINELLI ET AL.: "Very Low surface recombination velocities on p-type silicon wafers passivated with a dielectric with fixed negative charge", SOLAR ENERGY MATERIALS & SOLAR CELLS, vol. 90, 2006, pages 3438 - 3443, XP028002325, DOI: doi:10.1016/j.solmat.2006.04.014
SAINT-CAST ET AL.: "Very low surface recombination velocity on p-type c-Si by high-rate plasma-deposited aluminum oxide", APPLIED PHYSICS LETTERS, vol. 95, 2009, pages 151502, XP012126042, DOI: doi:10.1063/1.3250157
Attorney, Agent or Firm:
DE BEAUFORT, François-Xavier (Service DPI7, Quai d'Orsay Paris Cedex 07, FR)
Download PDF:
Claims:
CLAIMS

1 . Use of a composition comprising, for 100% of its total mass, at least 97% of an aluminium precursor of formula: AI(Ri)2(OR2) with Ri and R2 independently selected in the group consisting of methyl, ethyl, propyl, isopropyl and t-butyl; and :

- From 200 ppb to 5 ppm of Mo (Molybdenum);

- From 1000 ppb to 5 ppm of Fe (Iron);

- From 200 ppb to 5 ppm of Cu (Copper);

- From 200 ppb to 5 ppm of Ta (Tantalum);

for AI2O3 growth for silicon wafer based photovoltaic solar cells.

2. Use according to claim 1 , wherein said composition comprises also : - From 5000 ppb to 10 ppm of Ni (Nickel);

- From 2000 ppb to 10 ppm of Zn (Zinc);

- From 5000 ppb to 10 ppm of W (Tungsten).

3. Use of claim 1 or 2 wherein said aluminium precursor is AI(Me)2(OiPr).

4. Use of claim 1 or 2 wherein said composition comprises other metal impurities such as B (Boron), Ca (Calcium), K (Potassium), Cr (Chromium), Na (Sodium), Nb (Niobium), Ti (Titanium), Mn (Manganese), Co (Cobalt), Sn (Tin) in a quantity comprised between 0 and 20 ppm for each of them.

5. Use of a composition as defined in anyone of claims 1 to 4 for an AI2O3 thin film deposition method by ALD, PEALD, CVD, PECVD or MOCVD.

6. Use according to claim 5, wherein the deposition method is an ALD method.

7. Use according to anyone of claims 1 to 6 wherein the composition defined in one of claims 1 to 4 comprises from 97% to 99,4% of said aluminium precursor.

8. Film of AI2O3 obtained by the use as defined in anyone of claims 1 to 7.

9. Use of the film defined in Claim 8 for the passivation of photovoltaic devices, in particular for solar cells.

10. Use of the film defined in Claim 8 for the encapsulation of photovoltaic devices, in particular for solar cells.

Description:
Growth of AI2O3 thin films for photovoltaic applications

The present invention concerns a com position of aluminum source containing metal impurities, useful for the deposition of thin films in photovoltaic technologies, in particular for solar cells.

The photovoltaic effect is known since the end of the 19 th century. The principle consists in converting light energy into electricity. In the current ecological context, this is a promising solution to produce clean and renewable energy. One of the reasons for the slow development of photovoltaic electricity up to now is its lack of competitiveness compared to the traditional solutions such as coal, fossil fuels or nuclear based electricity. So the contribution of solar electricity as one significant component of the future energy mix is bounded to the capability to continue the cost per watt peak red uction race. To reach th is goa l , reduction of the manufacturing costs and improvement of cell's efficiency are two solutions.

Reduction of the manufacturing costs is addressed for example with thinner wafers usage to limit the impact of silicon feedstock price on the overall cell's cost (in 2010, cost split is 55 % for wafer - with 23 % for silicon feedstock, 13 % for silicon growth and 18 % for wafering step - and 45 % for cell processing), and in general with reduced raw materials consumption, including chemicals used during each step of the manufacturing. This manufacturing cost decrease is driven by manufacturing tools providers (the OEM - Original Equipment Manufacturers) and by material suppliers.

Improvement of photovoltaic cell's efficiency requires innovation activity that takes place in the Research laboratories. The passivation phenomenon related work carried out by academics may contribute to the enhancement of the photovoltaic cell's performance. Passivation of the cell's defects by, for example, hydrogen increases the efficiency of the cell by reducing the probability of hole- electron pairs' recombination at the surface and in the bulk of the silicon: the lower the number of defects into the material the higher the probability that charge carriers are collected. These recombinations take place on the front side of the solar cell as well as on the backside. To prevent the backside recombination, the deposition of a thin AI2O3 layer has been described in the literature (for example P.Vitanov and al. "Chemical Deposition of AI2O3 thin films on Si substrates" Thin Solid Films, 51 7 (2009), 6327-6330) as well as Hoex et al. "Silicon surface passivation by atomic layer deposited Al 2 0 3 " JOURNAL OF APPLIED PHYSICS 104, 044903 2008_demonstrated that very thin ALD- AI 2 O 3 layers (5-30 nm) can provide an excellent surface passivation for n-type and p-type silicon wafers after annealing at 425°C. This layer contributes to passivation in two ways: reduction of interface defects density and field-effect passivation thanks to high negative charge density at the silicon - alumina interface (G. Dingemans, R. Seguin, P. Engelhart, M. C. M. van de Sanden and W. M . M . Kessels; "silicon surface passivation by ultrathin AI2O3 films synthesized by thermal and plasma atomic layer deposition"; Phys. Status Solidi RRL4, N° 1 -2, 10-12 (2010)).

Recombination also occurs in the bulk of the silicon. Therefore, for an effective collection of the carriers, their diffusion length (average distance covered by a charge carrier before it gets recombined) or lifetime should be at least equal to the wafer thickness (V.G. Litovchenko, N. I. Klyuis, A.A. Evtukh, A.A. Efremov, A.V. Sarikov, et al. ; "Solar cells prepared on multicrystalline silicon subjected to new gettering and passivation treatments"; Solar Energy Materials and Solar Cells, 72 (2002), 343-351 ). Injection of atom ic hydrogen to passivate the defects and reduction of the amount of metallic impurities into the bulk are two major solutions to increase diffusion length. Several sources of contamination can be identified during the different manufacturing steps: from the silicon wafer production, including the silicon ingot or silicon casting steps, to the solar cell fabrication where lots of chem icals are used. The detrimental impact of some impurities such as copper or iron on the lifetime of charge carriers is known thanks to studies carried out for the sem iconductor industry. Figure 1 and Figure 2 detail conversion efficiency of p-type and n-type wafers based solar cells depending on concentration of selected metallic impurities.

Figure 1 shows the effect of metal impurity concentration on p-type Cz-Si based solar cells' conversion efficiency.

Figure 2 shows the effect of metal impurity concentration on n-type Cz-Si based solar cells' conversion efficiency.

The detrimental effect of iron in a boron doped (p-type) silicon wafer is described by Zoth et al ("Silicon contamination control by lifetime measurements"; in Analytical and Diagnostic Techniques for Semiconductor Materials, Devices and Processes (eds. Kolbesen et a l . ) , Th e Electrochemical Society proceedings, volume ECS PV 99-16 (1999)). During the high temperature firing step (about 800°C) required by most solar cell manufacturing processes, iron atoms, also those possibly and unintentionally deposited with the AI2O3, will be scattered uniformly into the silicon bulk due to diffusion phenomena. These atoms will be located on interstitial sites and some of them will combine reversibly with a boron atom. Injection of m inority carrier in silicon by photo excitation will dissociate the FeB pairs and therefore interstitial iron atoms will become recombination centers which are ten times stronger than FeB.

Iron will also form an oxide, when oxygen is present in the bulk. This oxide is an additional defect in the structure and then a recombination center.

Other metallic impurities such as copper and nickel will have an impact on the surface of the wafer where they will precipitate and form recombination centers. Titanium and molybdenum wi ll diffuse slowly into the bulk and form active recombination centers after high temperature firing.

To conclude, experts of silicon science consider that transition metals, more precisely iron, copper, tantalum or molybdenum, are particularly detrimental to the global performance of solar cells (A.A. Istratov, C. Flink, H. Hieslmair, S.A. McHugo, E.R. Weber; "Diffusion, solubility and gettring of copper in silicon"; Materials Science and engineering B72 (2000) 99-104).

It is easily understood that impurities can be introduced during the wafer processi ng itself or duri ng hand l ing . Duri ng so lar ce l ls' m anufacturi ng , contamination can result from usage of inappropriate grade chemicals. In a CVD process for example, these undesirable elements will be transferred in the gas phase with the main compounds and then incorporated into the layers to be deposited, more precisely, in the passivation layers.

Based on the above information resulting from extensive experience of the scientific community, it is obvious that AI2O3 deposition - for surface of silicon wafer based solar cells - performed by ALD, i.e Atomic Layer Deposition, or MOCVD, i.e Metal-Organic Chemical Vapor Deposition (Agostinelli et al. "Very Low surface recombination velocities on p-type silicon wafers passivated with a dielectric with fixed negative charge"; Solar Energy Materials & Solar Cells, 90 (2006), 3438- 3443), or PECVD, i.e. Plasma Enhanced Chemical Vapor Phase Deposition (Saint- Cast et al., Very low surface recombination velocity on p-type c-Si by high-rate plasma-deposited aluminum oxide, APPLIED PHYSICS LETTERS 95, 1 51502 _2009), or any other vapor-phase deposition technique, can induce incorporation of impurities in the deposited layer, these impurities are to be found originally in the aluminium source precursor used for such deposition processes. One could argue that the impurities would be segregated into the passivation alumina layer but this consideration does not take into account that each metallic impurity has a specific diffusion factor and that this diffusion phenomenon from the alumina layer to the silicon bulk increases with temperature. Therefore, the more contam inated the passivation layer, the more contam inated the silicon due to enhancement of diffusion during the mandatory annealing or other high temperature steps (see Figure 3).

Table 1 details typical impurity upper limits concentrations and requirements for an aluminium precursor to be used in the electronics industry (for example, MOCVD of epitaxial AIGaAs, AIGaN and InAIGaP films for applications in products such as high brightness LEDs and lasers and more recently ALD of aluminium oxide, which is the dielectric layer of choice in DRAM devices). This aluminium precursor is a precursor of formula AI(Ri) 2 (OR2) with Ri and R 2 independently selected in the group consisting of methyl, ethyl, propyl, isopropyl and t-butyl. These limits are based on process chemical guidelines. These guidelines give appropriate metal and general contam ination information to assist end users and chemical manufacturers. If the quantity of one of the impurities is superior to the present limit, the precursor does not meet the requirements for usage in the manufacturing of a semiconductor. The grade as defined in table 1 is named "semiconductor grade" in the present description.

Semiconductor grade

(ppbw)

B Boron < 100

Ca Calcium < 50

Cr Chromium < 50

Cu Copper < 150

Fe Iron < 80

K Potassium < 50

Na Sodium < 30

Ni Nickel < 50

Zn Zinc < 100

Ti Titanium < 50

Mn Manganese < 50

Co Cobalt < 50

Table 1 : Typical impurities concentration in semiconductor grade aluminum precursor of the invention

Data of Table 1 and the above discussion encourage a solar cell manufacturer to use "semiconductor grade" to avoid any risk of contamination and therefore any possible degradation of the cells' efficiency. A "non-semiconductor grade" precursor usage for deposition of AI2O3 surface passivation layer could be a non-negligible source of contamination with strong lifetime killers such as iron or copper.

AI2O3 is used in the semiconductor industry for several applications such as passivation and electrical insulation. For these applications, "semiconductor grade" aluminum precursor is commonly used to avoid any metallic impurities potentially generating undesirable phenomena such as leakage currents.

Nevertheless, the inventors of the present invention show that astonishingly the usage of a non-semiconductor grade aluminum precursor can be used for deposition of a surface passivation layer in crystalline silicon wafer based solar cells manufacturing as such non-semiconductor grade chemical usage does not induce any observable deterioration of the cell's performance (evaluated for example with the measurement of the lifetime carrier). Another application of the non-semiconductor grade could be the deposition of AI2O3 for encapsulation of photovoltaic devices. The present invention concerns a composition comprising, for 100% of its total mass, at least 97% of an aluminium precursor of formula: AI(Ri) 2 (OR2) with Ri and R 2 independently selected in the group consisting of methyl, ethyl, propyl, isopropyl and t-butyl; and :

From 200 ppb to 5 ppm of Mo (Molybdenum);

- From 1000 ppb to 5 ppm of Fe (Iron);

From 200 ppb to 5 ppm of Cu (Copper);

From 200 ppb to 5 ppm of Ta (Tantalum).

According to one embodiment of the invention, the composition comprises also:

From 5000 ppb to 10 ppm of Ni (Nickel);

From 2000 ppb to 10 ppm of Zn (Zinc);

From 5000 ppb to 10 ppm of W (Tungsten). According to another embodiment of the invention, the said aluminium precursor is AI(Me) 2 (OiPr).

Accord ing to another em bodi ment of the invention , the composition comprises other metal impurities such as B (Boron), Ca (Calcium), K (Potassium), Cr (Chromium), Na (Sodium), Nb (Niobium), Ti (Titanium), Mn (Manganese), Co (Cobalt), Sn (Tin) in a quantity comprised between 0 and 20 ppm for each of them.

Accord ing to one em bodiment, the invention concerns the use of a composition as above for a thin film deposition method by ALD, PEALD, CVD, PECVD or MOCVD.

According to one embodiment, the invention concerns the use as defined above, wherein the deposition method is an ALD method. According to one embodiment, the invention concerns the use as defined above, wherein the thin film is a film of AI2O3.

According to one embodiment, the invention concerns the use as defined above, for AI2O3 growth for photovoltaic applications.

According to one embodiment, the invention concerns the use as defined above, wherein the composition defined above comprises from 97% to 99,4% of said aluminium precursor. According to another embodiment, the invention concerns a film of AI2O3 obtained by the use as defined above.

According to one embodiment, the invention concerns the use of the film as defined above, for the passivation of photovoltaic devices, in particular for solar cells.

According to one embodiment, the invention concerns the use as defined above, for the encapsulation of photovoltaic devices, in particular for solar cells. From the background of the invention, the inventors understood that a

"semiconductor grade" aluminum precursor was mandatory for specific applications but they have concluded that surprisingly a "non semiconductor grade" aluminium precursor as defined in the present invention can be used for the purpose of depositing a surface passivation alumina layer for crystalline silicon wafer based photovoltaic solar cell. Actually, the examples detailed below show that contrary to what was expected, usage of "non semiconductor grade" aluminum precursor (chemical with enlarged tolerance on the amount of impurities than "semiconductor grade" one) is a suitable technical solution to achieve performance meeting the photovoltaic solar cells' expectations. Also for any photovoltaic devices, encapsulation issues can be addressed by Atomic Layer Deposition or any other suitable vapor phase deposition technique of AI2O3 deposited with "non- semiconductor" grade precursor. This chemical with such grade has the advantages of allowing deposition of AI2O3 layers for passivation by ALD, MOCVD, PECVD or any other derivative deposition methods without observable degradation of the performance of the solar cell (evaluated with minority carrier lifetime measurements). Furthermore, for encapsulation purposes, ALD or any other derivative deposition methods can be considered to fabricate an AI2O3 film from such "non-semiconductor grade" chemicals.

This novel intermediate grade precursor's usage is not solely limited to the deposition of a surface passivation layer for crystalline silicon wafer based photovoltaic solar cell but its benefit could be applied to other manufacturing steps in the photovoltaic industry such as encapsulation process. As well, such lower grade aluminium precursors could be used in various other application fields. The deposition of an AI2O3 thin film requires several steps:

a. Vaporizing the AI(Ri) 2 (OR2) source to form an aluminum source gas phase.

b. Introducing several precursors in vapor phase in a deposition device, wherein said precursors comprise said vaporized AI(Ri) 2 (OR2) source, and oxygen source, for example, selected from oxygen, carbon dioxide or nitrogen dioxide plasma, ozone, water.

In one embodiment of the invention, the vaporization of the new precursor "non-semiconductor grade" can be performed by introducing a gas in a canister containing the said AI(Ri) 2 (OR2) "non-semiconductor grade" source. The canister is preferably heated at a temperature which allows to vaporize the said source with a sufficient vapor pressure. The carrier gas can be selected from Ar, He, H 2 , N 2 or mixtures of them. The canister can for instance be heated at temperatures in the range of 15°C to 170°C. The temperature can be adjusted to control the amount of precursor in the gas phase.

In another embodiment of the invention, the said AI(Ri) 2 (OR 2 ) "non- semiconductor grade" source is fed in the liquid state to a vaporizer where it is vaporized.

In one embodiment of the invention, the pressure in the said canister is in the range from 133 mPa to 133.10 2 Pa. The said vaporized metal source is introduced into a reaction chamber where it is contacted to a solar cell substrate. The substrate can be covered with a material selected from the group Si, S1O2, SiN, SiON, and other silicon containing substrates and films and even other metal containing films. The substrate can be heated to sufficient temperature to obtain the desired film at sufficient growth rate and with desired physical state and composition. Typical temperature range from room temperature to 600°C and preferably from 150°C to 600°C. Preferably the temperature is lower or equal to 450°C. The pressure in the reaction chamber is controlled to obtain the desired metal containing film at sufficient growth rate. The typical pressures range from 133 mPa to 133.10 2 Pa or higher.

In one embodiment of the invention, the said AI(Ri) 2 (OR2) "non- semiconductor grade" source described above is mixed to one or more reactant species prior to the reaction chamber.

In one embodiment of the invention, the said AI(Ri) 2 (OR2) "non- semiconductor grade" source described above is mixed to one or more reactant species in the reaction chamber.

In another embodiment of the invention, the said AI(Ri) 2 (OR2) non semiconductor grade source and the reactant species are introduced sequentially in the reaction chamber. The AI(Ri) 2 (OR2) "non-semiconductor grade" source and the reactant species can be introduced simultaneously (chemical vapor deposition), sequentially (atomic layer deposition) or different combinations (one example is to introduce the reactant species (one example could be oxygen) continuously and to introduce AI(Ri)2(OR 2 ) "non-semiconductor grade" source by pulse (pulsed- chemical vapor deposition).

In one embodiment of the invention, the reactant species are decomposed into radicals in a plasma system prior to entering in the reaction chamber.

In one embodiment of the invention the said reactant species include an oxygen source which is selected from oxygen (0 2 ), oxygen-containing radicals (for instance 0 or OH ), for instance generated by an oxygen (O2), carbon dioxide (CO2) or nitrogen dioxide (NO2) plasma, ozone (O3), moisture (H 2 O) and H2O2.

In one embodiment of the invention, the said AI(Ri) 2 (OR2) "non- semiconductor grade" source is used for atomic layer deposition of AI2O3 films. The said aluminum source and the reactant species are introduced sequentially in the reaction chamber (atomic layer deposition). The reactor pressure is selected in the range from 133 mPa to 133.10 2 Pa. Preferably the reactor pressure is comprised between 133 Pa and 1330 Pa. A purge gas is introduced between the metal source pulse and the reactant species pulse. The purge gas can be selected from the group consisting of N 2 , Ar, He. The aluminum source, purge gas and reactant species pulse duration is comprised between 1 ms and 10 seconds. The pulse duration is preferably about 10 ms.

Now, the present invention will be explained with reference to examples. 1 ) Al 2 0 3 thin film deposited by thermal ALD from an AI(Ri) 2 (OR 2 ) grade source according to the present invention.

AI(Me) 2 (OiPr) (DMAI) grade according to the present invention is stored into a canister. The canister is maintained at a temperature of 60°C. DMAI vapour is drawn but nitrogen is optionally used as a carrier gas at a flow of about for example 50 seem. The silicon substrate is heated at 200°C. Water is used as oxygen source. The alum inum precursor is sequentially introduced into the reaction chamber with water. During the first step, a pulse of AI(Me) 2 (OiPr) is introduced during 50 ms to 500ms, followed by an optional 3.5 s Argon purge. A pulse of water is then introduced into the reaction chamber during 20 ms, followed by the optional 3.5 s Argon purge. This first sequence is then repeated 1000 times. An Al 2 03 film meeting the electrical properties requirements (charge carrier lifetime of silicon substrate) for photovoltaic applications is obtained. Such a good result shows that the usage of a "non-semiconductor grade" is possible for deposition of an Al 2 03 layer having proficient passivation properties.

2) Al 2 03 thin film deposited by plasma ALD from an AI(Ri) 2 (OR 2 ) grade source according to the present invention. AI(Me) 2 (OiPr) (DMAI) grade according to the present invention is stored into a canister. The canister is maintained at a temperature of about 60°C. DMAI vapour is drawn but nitrogen is optionally used as a carrier gas at a flow of about for example 50 seem. Oxygen gas (50 seem) and Argon gas (50sccm) are continuously introduced in the chamber. The silicon substrate is heated at 200°C. Oxygen plasma is used as oxygen source. The aluminum precursor is sequentially introduced into the reaction chamber. During the first step, a pulse of AI(Me) 2 (OiPr) is introduced during 50 ms to 500ms, followed by an optional 3.5 s Argon purge. A plasma is generated for 3s to generate oxygen plasma and is followed by the optional 3.5 s Argon purge. This first sequence is then repeated 1000 times. An AI2O3 film meeting the electrical properties requirements (charge carrier lifetime of silicon substrate) for photovoltaic applications is obtained. Such a good result shows that the usage of a "non-semiconductor grade" is possible for deposition of an AI2O3 layer having proficient passivation properties.

3) Deposition tests were performed with a "semiconductor grade" (the highest grade available, see Table 1 for some typical impurity levels) and a "non- semiconductor grade" (see Table 2) AI(Me) 2 (OiPr) samples. Resulting recombination velocities of minority carriers (before and after a 400°C annealing step) are given (see Figure 4) for 20 nm thick AI2O3 layer deposited by thermal ALD and plasma ALD using a "semiconductor grade" (highest purity) and a "non- semiconductor grade" according to the present invention. Figure 4 shows that the results of the AI2O3 layer deposited from AI(Me) 2 (OiPr) by "non-semiconductor grade" according to the present invention are equivalent to the results of the AI2O3 layer deposited from AI(Me) 2 (OiPr) by "semiconductor grade".

Non- Semiconductor grade of

the invention (ppbw)

B Boron 5,000

Ca Calcium 5,000

Cr Chromium 5,000

Cu Copper 5,000

Fe Iron 3,000

K Potassium 10,000

Na Sodium 6,000

Ni Nickel 10,000

Zn Zinc 10,000 Ti Titanium 5,000

Sn Tin 5,000

Mo Molybdenum 2,000

Ta Tantalum 2,000

W Tungsten 10,000

Table 2: Typical impurities concentration in non-semiconductor grade

AI(Me) 2 (OiPr) precursor

Table 3 Typical impurities concentration in non-semiconductor grade AI(Me) 2 (OiPr) precursor

Taking into consideration the background of the invention, the conclusion of these tests is very surprising and it opens up a new possibility for manufacturing of crystalline silicon wafer based photovoltaic solar cells thanks to this innovative "non-semiconductor grade" AI(Ri) 2 (OR 2 ) solution for deposition of surface passivation alumina layer. Also for any photovoltaic devices, encapsulation issues can be addressed by Atomic Layer Deposition of AI2O3 deposited with "non- semiconductor grade" AI(Ri) 2 (OR2).

It was demonstrated that the deposition of AI 2 0 3 for passivation can be performed from a "non-semiconductor grade" of AI(Ri) 2 (OR2) such as AI(Me) 2 (OiPr) (DMAI) as aluminum source without observable degradation of the crystalline silicon wafer based photovoltaic solar cells' performance; and more precisely without harming the passivation capability of the surface alumina layer.