Synthesis and Characterization of Alkylamine-Functionalized Si(111

Subscriber access provided by GRIFFITH UNIVERSITY

Article

Synthesis and Characterization of Alkylamine-Functionalized Si(111) for Perovskite Adhesion with Minimal Interfacial Oxidation or Electronic Defects Alexander D Carl, Roghi E Kalan, John David Obayemi, Martiale Gaetan Zebaze Kana, Winston Oluwole Soboyejo, and Ronald L. Grimm ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b07117 • Publication Date (Web): 07 Sep 2017 Downloaded from http://pubs.acs.org on September 9, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Synthesis and Characterization of Alkylamine-Functionalized Si(111) for Perovskite Adhesion With Minimal Interfacial Oxidation or Electronic Defects

Alexander D. Carl,1 Roghi E. Kalan,1,2 John David Obayemi,3 Martiale Gaetan Zebaze Kana,3,4 Winston Oluwole Soboyejo,3 and Ronald L. Grimm1*

1

Department of Chemistry and Biochemistry; Life Science and Bioengineering Center; Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609 2

Now at AR Metallizing Ltd.; 24 Forge Park, Franklin, Massachusetts 02038

Department of Mechanical Engineering; Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609 3

Department of Mechanical Engineering; Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609 3

Department of Materials Science and Engineering; Kwara State University, PMB 1531, Malete, Nigeria

4

* Author to whom correspondence should be addressed: [email protected]

Keywords: Tandem-junction solar cell, perovskite, silicon, photoelectrochemistry, x-ray photoelectron spectroscopy

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 54

Abstract We investigated synthetic strategies for the functionalization of Si(111) surfaces with organic species containing amine moieties. We employed the functionalized surfaces to chemically “glue” perovskites to silicon with efficient electron transfer, and minimal oxidation leading to deleterious recombination at the silicon substrate. A two-step halogenation-alkylation reaction produced a mixed allyl-methyl monolayer on Si(111). Subsequent reactions utilized multiple methods of brominating at the allyl double bond including reaction with HBr in acetic acid, HBr in THF and molecular bromine in dichloromethane. Reaction with ammonia in methanol effected conversion of the bromide to the amine. X-ray Photoelectron Spectroscopy (XPS) quantified chemical states

and

coverages,

transient-microwave

photoconductivity

ascertained

photogenerated carrier lifetimes, Atomic Force Microscopy (AFM) quantified perovskite-silicon adhesion, and nonaqueous photoelectrochemistry explored solarenergy-conversion performance.

The HBr bromination followed by the amination

yielded a surface with ~10% amine sites on the Si(111) with minimal oxide and surface recombination velocity values below 120 cm s–1 following extended exposures to air. Importantly,

conversion

of

amine

sites

to

ammonium

and

deposition

of

methylammonium lead halide via spin coating and annealing did not degrade carrier lifetimes. AFM experiments quantified adhesion between perovskite films and alkylammonium-functionalized

or

native-oxide

silicon


surfaces.

Adhesion

Page 3 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


forces/interactions between the perovskite and the alkylammonium-functionalized film was comparable to the interaction between the perovskite and native-oxide silicon surface. Photoelectrochemistry of perovskite thin films on alkylamine-functionalized n+-Si showed significantly higher Voc than n+-Si with a native oxide when in contact with a nonaqueous ferrocene+/0 redox couple. We discuss the present results in the context of utilizing molecular organic recognition to attach perovskites to silicon utilizing organic linkers so as to inexpensively modify silicon for future tandem-junction photovoltaics.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. Introduction Silicon based photovoltaics have dominated research and commercial implementation of solar energy devices since Bell labs unveiled the first practical cell in 1954.1 Present efficiencies for single junction, crystalline silicon solar cells, under non-concentrated light, exceed 25%.2 As efficiency approaches the Shockley–Queisser limit,3-4 the increases in efficiency between iterations decrease. In contrast to single-junction, monolithic cells, tandem-junction cells are typically employed for their high theoretical energyconversion efficiencies.5 Approaches to tandem-junction designs have typically involved sequential vapor deposition of III-V materials to yield, double-, triple-, or quadruple-junction cells. An alternative to multijunction III-V photovoltaics capitalizes on the 60 years of research on silicon to pair it with a larger gap material that may be inexpensively deposited on the silicon surface. To match currents, the 1.12 eV energy band gap in silicon could be paired with a 1.5–2.0 eV gap material for modest gains or a 1.7–1.8 eV gap material for maximized efficiency.6-8

Perovskites represent a particularly

compelling class of materials for pairing with silicon. Recent research on organicinorganic hybrid perovskites has demonstrated low-cost, versatile cells, with NRELcertified efficiencies recently exceeding 22%.9 A singly-charged organo-cation such as methylammonium or formamidinium; a metal cation such as Pb2+, Sn2+, or Bi3+; and a singly-charged halide anion typically comprise solar-relevant perovskites in A1+ B2+ X–1 3 or


Page 4 of 54

Page 5 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3+ –1 A1+ stoichiometries.10 Compositional variations enables both stabilization and 3 B2 X 9

band gap tuning for single-junction or multi-junction implementations.11 Critical to the formation of efficient perovskite-silicon heterojunction cells is an interconnecting recombination layer that demonstrates carrier-selective transport, strong adhesion to both the perovskite and silicon materials, and minimal deleterious state formation. Zinc, titanium, and tin oxides are typical candidates for the selective transport layer.6-7 However, metal oxides demonstrate limited energy level tunability that may not necessarily match the “ideal” perovskite band edges, and the deposition of metal oxides may ultimately lead to oxygen diffusion into the silicon and a concomitant creation of deleterious recombination sites.12 Organic interconnecting layers with a wide energy level tunability that may be deposited via near-ambient temperature and pressure conditions may yield a desirable alternative to metal oxides. The organic layer should effectively passivate the silicon surface against atmospheric oxidation as well as degradation that may occur during perovskite processing steps. Such interconnecting layers may be covalently attached to silicon for long-term stability, and exhibit terminal functionality that is “recognized” by the perovskite for strong adhesion to the interconnecting layer. Investigation of such organic interconnecting layers and their silicon passivation properties motivates the present study. Herein, we investigate the passivation and functionalization of monolayers on Si(111) to introduce terminal ammonium functionality for the purpose of chemically



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 54

“gluing” perovskites to the silicon and characterize the adhesion of the perovskite film to silicon and its resulting photoelectrochemical performance. We utilize a two-step halogenation-alkylation procedure to initially functionalize the silicon surface.13 Alternative functionalization strategies with broad functional group tolerance exist,14-19 however the two-step halogenation-alkylation demonstrates excellent passivation properties with low surface-recombination velocities following prolonged air exposures.20 Mixed-monolayers utilizing methyl and allyl groups enables a high coverage of silicon sites for effective passivation with a chemical handle for further functionality.21-26 We explore functionalization strategies to brominate the double bond on a mixed allyl/methyl-functionalized Si(111) surface prepared by a

two-step

halogenation-alkylation, and subsequently aminate the alkylbromide site for perovskite deposition.

The ultimate goal involves the production of an alkylammonium-

functionalized silicon surface that demonstrates minimal oxidation, strong perovskitesilicon adhesion, and good electronic properties. X-ray photoelectron spectroscopy (XPS) experiments and substrate-overlayer-coverage models quantified interfacial silicon oxidation and fractional surface coverage of bromine and nitrogen species. As the morphology and arrangement of bromine and ammonium groups on the surface is not quantified by XPS, we include a sensitivity study to consider how variations in values employed in the substrate-overlayer equations may affect our interpretation of the photoelectron results. Infrared reflection-absorption spectroscopy (IRRAS) verified


Page 7 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the presence of allyl and of methyl groups on the mixed monolayer surfaces. Timeresolved

microwave

photoconductivity

spectroscopy

(TRMPS)

measurements

ascertained the carrier recombination lifetimes and surface recombination velocity (SRV) values of photogenerated carriers in the silicon. While we are not concerned with photogenerated carrier lifetimes in bulk silicon as it directly relates to tandem-junction solar cells per se, SRV values are an important proxy for the density of defect states at the silicon surface that promote deleterious recombination at the silicon/perovskite interface. Adhesion force techniques determined the adhesion force between MAPbI3 films and silicon substrates. Photoelectrochemistry measured the current density– potential, J–E, curves of perovskite thin films deposited on degenerately doped n+–Si which served as the ohmic back contact to the perovskite. Photoelectrochemical experiments contrasted the electron transfer and recombination of charge-carriers at the back contact as a function of the respective silicon surface chemistry.

2. Experimental Section 2.1. Materials and chemicals The experiments utilized single-side polished, 525 ± 25 µm thick, 1–5 Ω cm resistivity, phosphorous-doped

n–Si(111)

photoelectron spectroscopy.

(EL-CAT

Inc.,

Ridgefield

Park,

NJ)

for

x-ray

The photoconductivity experiments used double-side

polished, 525 ± 25 µm thick, >3000 Ω cm resistivity, intrinsic silicon, i–Si(111), (EL-CAT



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 54

Inc.). Photoelectrochemistry and AFM experiments employed single-sided polished, 525 ± 15 µm thick, degenerately arsenic doped, ≤0.006 Ω cm resistivity wafers (Wafer Works Corp.) that are termed n+-Si(111) below. Reagents for etching and cleaning silicon included the RCA Standard Clean–1 (SC–1 or RCA–1) solution, the RCA Standard Clean–2 (SC–2 or RCA–2) solution, aqueous hydrogen fluoride, and aqueous ammonium fluoride.

*Warning, aqueous

hydrogen fluoride is an acute poison that is toxic at even small amounts and limited exposures*. A Millipore Milli-Q system provided 18 MΩ cm resistivity water for all water requirements. A solution containing five volume-parts water, one part H2O2(aq) (35 wt %, Alfa Aesar) and one part NH4OH(aq) (28% NH3, Alfa Aesar) comprised the SC–1 solution. The SC–2 solution consisted of 5 volume-parts water, one part hydrogen peroxide, and one part hydrochloric acid (12 M, ACS grade, Alfa Aesar). The HF(aq) was diluted to 6 M from a commercial stock solution (49%, electronic grade, Transene Company Inc., Danvers, Massachusetts). The ammonium fluoride solution (40% in water, electronic grade, Transene) was sparged with argon (ultra-high purity, Airgas) for 30 minutes prior to use. A stream of argon dried the silicon samples as needed. Chemicals employed for the two-step halogenation-alkylation functionalization of silicon surfaces included liquid bromine (99.8%, Alfa Aesar) that was degassed via three freeze-pump-thaw cycles on a diffusion-pump-equipped, argon-purged Schlenk line with a base pressure below 1 × 10–3 Torr and stored over degassed P2O5 (99.9925%,


Page 9 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Alfa Aesar). Tetrahydrofuran (THF, 99.8%, anhydrous, Alfa Aesar,) was degassed prior to experiments via three freeze-pump-thaw cycles on the Schlenk line and stored over activated 3 Å molecular sieves (Alfa Aesar). Methylmagnesium chloride (CH3MgCl, 3 M in THF, Alfa Aesar) and allylmagnesium chloride (CH2CHCH2MgCl, 2 M in THF, Sigma Aldrich) were diluted to 1 M in anhydrous THF during experiments. Molecular sieves further maintained “dry” methanol (99.9% min, semiconductor grade, Alfa Aesar) and glacial acetic acid (AcOH, 99.7%, Alfa Aesar). Benzoyl Peroxide (99.7% dry mass, wet with 25% water, Alfa Aesar) was dried under vacuum. Chemicals utilized for the functionalization of the allyl double bond included dichloromethane (DCM, 99.5%, Pharmco-Apper), hydrogen bromide (33 wt % in glacial acetic acid, Alfa Aesar), and glacial acetic acid. Dichloromethane and glacial acetic acid were dried over activated molecular sieves and the Br2/DCM and HBr/AcOH solutions were degassed immediately preceding use via three freeze-pump-thaw cycles. Chemicals employed for the amination of the alkylbromide included ammonia (2 M in methanol, Alfa Aesar) and sodium hydroxide (98.5%, Acros). Three freeze-pumpthaw cycles removed dissolved oxygen from the ammonia solution. Chemicals for the deposition of methylammonium lead iodide included HCl(aq) (11.6 M, Alfa Aesar, as received), lead iodide (lead(II) iodide, 99.9985% metals basis, Alfa Aesar) and methylammonium iodide. Reaction of 10 mL CH3NH2(aq) (21 M, Alfa Aesar, as received) with 20 mL of HI(aq) (10 M, Alfa Aesar, as received) at 25 °C with



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 54

stirring for 2 h, followed by solvent removal by rotary evaporation produced methylammonium iodide. Washing the crude methyl ammonium iodide product three times with diethyl ether (99+%, spectrophotometric grade, inhibitor free, Alfa Aesar, as received) followed by solvent removal on the Schlenk line afforded ~15 g of methylammonium

iodide.

Additional

solvents for

the

spin

deposition of

methylammonium lead iodide included N,N-dimethylformamide (DMF, ≥99.8%, Certified ACS, Fisher Chemical). Chemicals for the photoelectrochemistry included lithium perchlorate (battery grade, Sigma-Aldrich), ferrocene (Cp2Fe, bis(cyclopentadienyl)iron(II), 99%, SigmaAldrich),

ferrocenium

(Cp2FePF6,

hexafluorophosphate

bis(cyclopentadienyl)iron(III)hexafluorophosphate,

97%,

Aldrich),

and

1,1,2,2-

tetrachloroethane (98+%, Alfa Aesar). Ferrocene was purified by vacuum sublimation while ferrocenium hexafluorophosphate was purified via recrystallization and further degassed

under

vacuum.

Lithium

perchlorate

was

used

as

received.

Photoelectrochemical experiments utilized a ferrocene+/0 solution consisting of 100 µM of ferrocene, ~5 µM ferrocenium, and 1 M LiClO4 in “dry” tetrachloroethane. The ferrocene+/0 solution was stored over activated molecular sieves for several days prior to use. A series of freeze-pump-thaw cycles on the Schlenk line and storage over activated molecular sieves prepared the tetrachloroethane at least one week prior to use.


Page 11 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Dicing, oxidation, and etching steps preceded wafer functionalization. The SSP n-Si(111) wafers for photoelectron spectroscopy experiments were diced into ~0.3 cm2 pieces, while wafers for all other experiments were diced in to 2–3 cm2 pieces. Initial cleaning consisted of a 10-minute exposure to the SC–1 solution at 77 °C, a water rinse, a 10 s submersion in 6 M HF(aq), copious rinsing in water, and drying under argon. Samples subsequently cleaned via a 10 min submersion in the SC–2 solution at 77 °C, rinsing in water, a 10 s submersion in 6 M HF(aq), a copious water rinse, and drying under argon. Following the second HF exposure, a four-minute immersion in argonsparged NH4F(aq), rinsing with H2O, and drying under argon effected large hydrogenterminated terraces on the Si(111) surface. Following the ammonium fluoride etch, samples were immediately transferred to an evacuated reaction vessel.

2.2. Wafer preparation and functionalization A two-step halogenation-alkylation reaction functionalized the hydrogen-terminated Si(111) samples. An air-free reaction vessel connected to the diffusion-pumped Schlenk line contained silicon wafers in individual wells to minimize scratching from adjacent wafers.

Three pump-argon-purge cycles effected atmospheric O2(g) removal.

The

halogenation step utilized gas-phase bromine delivered via the vapor headspace from liquid Br2. Prior to Br2(g) exposure, the raising the pressure in the Schlenk line and the sample reaction vessel to ~500 mTorr of argon minimized bumping in the liquid



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 54

bromine source flask. The sample reaction vessel was opened to the liquid bromine flask for 1 s allowing Br2 vapor to fill the vessel. A two-minute exposure brominated the silicon surfaces in the reaction vessel, which was subsequently evacuated and filled with an argon ambient to atmospheric pressure. Unrelated to the ongoing functionalization methods, three brominated Si(111) samples were expeditiously transferred to the XPS with nominal ambient air exposure. XP spectra quantified the Br 3d and Si 2p regions of the brominated Si(111) samples to test and validate the substrate-overlayer model that is employed below to ascribe surface coverages from XPS data.

The Supporting Information contains the photoelectron spectra and

interpretation for bromine-terminated Si(111). Grignard chemistry functionalized the bromine-terminated Si(111).

Cannula

transfer filled each well of the silicon reaction vessel with 3–6 mL of a THF solution of 0.2 M CH2CHCH2MgCl and 0.8 M CH3MgCl under argon to keep each wafer submerged throughout the reaction period. Separately, a solution of THF containing only 3-6 mL of 0.8 M CH3MgCl prepared methyl-terminated Si(111) surfaces for adhesion studies that were otherwise processed in a similar manner to the mixed allyl/methyl-terminated silicon samples. An oil bath maintained at 72 ± 1 °C heated the reaction vessel throughout the three-hour reaction period. Following Grignard reactions, 3–6 mL THF was added via an air-free syringe. The reaction vessel was opened to ambient atmosphere, and each silicon wafer was expeditiously removed,


Page 13 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


placed in a centrifuge tube, and submerged in THF for initial sonication. The samples were sonicated in 10 minute, sequential steps in THF, methanol, and argon-sparged water. The sparging step minimized dissolved carbon dioxide that could otherwise acidify the water and deleteriously react with the freshly prepared double bond. Bromination of the double bond was initiated directly following alkylation as reports note that exposure to the air negatively affects the reactivity of the surface toward bromination.27 We explored three methods to brominate the carbon-carbon double bond for ultimate amination with minimal silicon oxidation and minimal degradation of electronic properties. All bromination reactions were completed in an air-free reaction vessel connected to the Schlenk line. Method A involved bromination of the double bond with a 1.2 M HBr solution consisting of 9 volume parts THF to 2 parts glacial acetic acid (formed by a dilution of 1 volume part stock solution into 4 parts THF). A small amount (

Synthesis and Characterization of Alkylamine-Functionalized Si(111

Recommend Documents