Fabrication, Pressure Testing, and Nanopore Formation of Single

Jun 1, 2017 - Single-layer graphene (SLG) membranes have great promise as ultrahigh flux, high selectivity membranes for gas mixture separations due t...
0 downloads 7 Views 2MB Size
Subscriber access provided by Binghamton University | Libraries

Article

Fabrication, Pressure Testing and Nanopore Formation of Single Layer Graphene Membranes Kumar Varoon Agrawal, Jesse D Benck, Zhe Yuan, Rahul Prasanna Misra, Ananth Govind Rajan, Yannick Eatmon, Suneet Kale, Ximo S Chu, Duo O Li, Chuncheng Gong, Jamie H. Warner, Qing Hua Wang, Daniel Blankschtein, and Michael S. Strano J. Phys. Chem. C, Just Accepted Manuscript • Publication Date (Web): 01 Jun 2017 Downloaded from http://pubs.acs.org on June 4, 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.

The Journal of Physical Chemistry C 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 29

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 Journal of Physical Chemistry

Fabrication, Pressure Testing and Nanopore Formation of Single Layer Graphene Membranes

Kumar Varoon Agrawal1,*‡ Jesse D. Benck1,* Zhe Yuan1, Rahul Prasanna Misra1, Ananth Govind Rajan1, Yannick Eatmon1, Suneet Kale2, Ximo S. Chu2, Duo O. Li2, Chuncheng Gong3, Jamie Warner3, Qing Hua Wang2, Daniel Blankschtein1, Michael S. Strano1†

1

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

2

Materials Science and Engineering, School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, AZ, USA 3

Department of Materials, University of Oxford, Oxford, United Kingdom * Equally contributing authors †



Corresponding author’s email address: [email protected]

Current address: Institute of Chemical Sciences and Engineering, École Polytechnique Fédérale de Lausanne, Route Cantonale, 1015 Lausanne, Switzerland

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 29

Abstract Single-layer graphene (SLG) membranes have great promise as ultrahigh flux, high selectivity membranes for gas mixture separations due to their single atom thickness. It remains a central question whether SLG membranes of a requisite area can exist under an imposed pressure drop and temperatures needed for industrial gas separation. An additional challenge is the development of techniques to perforate or otherwise control the porosity in graphene membranes to impart molecularly sized pores, the size regime predicted to produce high gas separation factors. Herein, we report fabrication, pressure testing, temperature cycling, and gas permeance measurements through free-standing, low defect density SLG membranes. Our measurements demonstrate the remarkable chemical and mechanical stability of these 5 µm diameter suspended SLG membranes, which remain intact over weeks of testing at pressure differentials of > 0.5 bar, repeated temperature cycling from 25 °C to 200 °C, and exposure to 15% mole ozone for up to 3 min. These membranes act as molecularly impermeable barriers, with very low or near negligible background permeance. We also demonstrate a 1077 °C temperature O2 etching technique to create nanopores on the order of ~1 nm diameter as imaged by scanning tunneling microscopy, although transport through such pores has not yet been successfully measured. Overall, these results represent an important advancement that will enable graphene gas separation membranes to be fabricated, tested and modified in situ while maintaining remarkable mechanical and thermal stability.

2 ACS Paragon Plus Environment

Page 3 of 29

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 Journal of Physical Chemistry

Introduction As the thinnest possible molecularly impermeable barrier, graphene could be a transformative gas separation membrane.1-10 Its single atom thickness gives it the potential to achieve gas fluxes orders of magnitude higher than polymer membranes.3-5 Membrane-based separations are attractive as replacements for thermal separations such as distillation because they could reduce overall energy consumption by up to 90%.11-14 However, the performance of conventional membrane technologies precludes their usage in many large-scale industrial separations. 11-13, 15 Gas permeation through sub-nanometer nanopores in the graphene lattice can be an activated process, and differences in the activation energy for different gas species may grant graphene membranes extremely high mixture separation selectivity.3-5,

16-18

The high

mechanical strength19 and chemical stability1 of the graphene lattice are also attractive for membrane applications. Graphene membranes with ultrahigh flux and selectivity could substantially decrease the area needed to separate a given volume of a gas mixture, providing a new solution to the problem of membrane scale up, a longstanding challenge in the field.14, 20-21 For example, a graphene membrane with 10-3 mol m-2 s-1 Pa-1 H2 permeance, a value which is likely attainable,6, 17 could achieve 106 standard cubic meters per day H2 flux through only 5 m2 membrane area at 1 bar pressure differential. A typical polymer membrane would need to be 5 × 103 to 5 × 106 m2 in area to achieve the same gas throughput under these conditions. Previous theoretical and computational studies have predicted that graphene membranes with nanometer sized pores could have high flux and selectivity for gas mixture separations.16-17, 22-41

Experimentally, single gas transport through graphene membranes with tears or rips42 and

through capillaries between sheets of multilayered graphene or graphene oxide has been measured.7-8 Bunch and coworkers recently reported size-selective gas permeance through

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 4 of 29

suspended graphene membranes measuring one gas species at a time.10 Park, et al. studied gas mixture permeation through graphene membranes with pores too large for activated translocation.6 In spite of these important efforts, there are several unanswered questions that must be addressed to enable a more complete understanding of gas permeation through graphene membranes. Measuring the temperature dependence of transport behavior is an important and unexplored strategy for understanding gas permeation mechanisms,18 but so far, the stability of suspended graphene membranes under repeated temperature cycling has not been established. In addition, the development of improved experimental methodologies to directly measure the trans-membrane flux of each species in a gas mixture rather than only single gas fluxes remains a significant challenge. Competitive adsorption and diffusion may influence mixture separations for this type of membrane,17-18,

43-44

and as yet these phenomena have not been extensively

investigated. Another important challenge is the development of improved techniques to perforate graphene membranes with pores of ~1 nm diameter or smaller, the size range predicted to be necessary to observe activated transport for small gas molecules such as CH4 and CO2.3-5, 18, 25

Addressing these challenges will be critical for developing an improved understanding of the

relationship between atomic-scale pore structure and gas permeance, enabling the design of graphene membranes with properties tailored to achieve high flux and separation factors. In this work, we investigate the thermal and chemical stability of suspended single layer graphene membranes using a custom-built gas permeation module with an on-line mass spectrometer configured to directly measure gas flux through the membranes. For the first time, we show that these graphene membranes exhibit exceptional strength and stability, withstanding pressure differentials (∆P) up to 0.5 bar, repeated temperature cycling from 25 °C to 200 °C, and exposure to 15% mole ozone without rupturing over many days of testing, results which have

4 ACS Paragon Plus Environment

Page 5 of 29

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 Journal of Physical Chemistry

previously never been experimentally demonstrated. The flux of all studied gas species, including CO2, CH4, SF6, H2, He, and Kr, is lower than the detection limit of our technique. When a membrane eventually ruptures after extensive testing, we measure a significant steady state flux of the feed gases through the open hole of the punctured graphene over the metal foil support. Our data provide preliminary evidence that ozone treatment can create pores in the membranes, providing a promising avenue for future study. We also report a new method of oxygen etching of the graphene lattice to form pores on the order of ~1 nm diameter as confirmed by scanning tunneling microscopy (STM). There have been very few studies in the literature describing oxygen etching techniques to create molecularly-sized pores,45 but the development of such techniques will be very important because nanopores in this size range may have the ability to achieve size-selective gas separation. Our results, which reveal the stability of our graphene membranes, demonstrate the effectiveness gas mixture permeation measurement technique, and provide a new technique for graphene nanopore formation, are critical to establish as a first step towards studying gas permeance through graphene membranes. As such, this work represents important progress towards understanding molecular sieving separation through nanoporous graphene materials.

Methods Electropolishing Cu foil substrates for graphene synthesis Copper foils (25 µm thick, 6.3 cm long, 2.5 cm wide, 99.999% purity, Alfa-Aesar) were electropolished in an electrolyte solution made by mixing 100 mL water, 10 mL isopropanol, 50 mL ethanol, 50 mL orthophosphoric acid, and 1 g urea. A piece of Cu foil was used as the anode in the electrolytic cell, and another piece of 25 µm thick Cu foil (30 cm x 2.5 cm, 99.8% purity,

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 29

Alfa-Aesar) was folded to 6 cm x 2.5 cm and used as the cathode. The distance between the anode and cathode was kept at 3 cm to prevent H2 bubbles generated at the cathode from reaching the anode. Electropolishing was carried out for 3 minutes at an applied voltage of 3.5 V vs. the counter electrode. The resulting current was steady at 0.5 – 0.6 A. Thereafter, the Cu foil was rinsed with ethanol and repeatedly sonicated in ethanol to remove any remaining oxidized copper particles from the surface. Subsequently, the foil was rinsed with water and isopropanol, and cut to 5 cm by 2 cm coupons for the graphene synthesis.

Controlled pressure chemical vapor deposition (CPCVD) graphene synthesis Graphene synthesis was carried out using controlled pressure chemical vapor deposition (CPCVD) following the procedure from Tour and coworkers46 with minor modifications. The reactor setup in illustrated in Supporting Figure S1. The electropolished Cu foil was suspended on a homemade 3 mm thick Cu frame (Supporting Figure S2). The frame provided mechanical support to the thin foil during the CVD near the melting point of Cu, and enabled flat, wrinklefree cut-out of the Cu/graphene coupons after the CPCVD synthesis. To begin the synthesis, the Cu foil was placed on top of the frame at the center of the reactor (quartz tube with an outer diameter of 25 mm, heated with MTI vacuum tube furnace, OTF-1200X-S) while maintaining a steady H2 flow of 500 sccm using a MKS GE50A mass flow controller (MFC). Following this, the reactor was evacuated while maintaining the H2 flow (500 sccm) to purge out the atmospheric gases. The lowest pressure obtained during the evacuation (with H2 flow) was 2 torr (MKS 722B Baratron capacitance manometer). Next, the H2 flow was lowered to 70 sccm, and using a metering valve (SS-4BMG, Swagelok) downstream of the Cu foil (Supporting Figure S1), the pressure in the reactor was raised to 1.5 bar. Annealing of the Cu foil, a step necessary to

6 ACS Paragon Plus Environment

Page 7 of 29

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 Journal of Physical Chemistry

create a smooth Cu surface and reduce the graphene nucleation density,46-47 was carried out by holding Cu foil in the H2 atmosphere at 1077 °C for 15 – 20 hours. After annealing, to initiate nucleation and growth of graphene, the reactor pressure was reduced to 108 torr, and 0.1 sccm CH4 was flowed into the reactor using a MKS GE50A MFC along with the 70 sccm H2 stream. After 6 – 8 hours growth time, the CH4 flow was switched off, and the quartz tube was pushed out of the furnace quickly (after ~2 seconds) to enable fast cooling, thereby terminating graphene growth. The cooling was aided by blowing air over the outside of the quartz tube using a fan (12 cm AC fan, StarTech). After 1 hour, the reactor pressure was raised to 1 bar, and the Cu foil was taken out of the reactor.

Fabrication of the tungsten foil support A tungsten foil (50 µm thick, 1 cm long, 1 cm wide, Alfa Aesar) was chosen as the membrane support due to its mechanical and thermal stability. A single cylindrical hole with a diameter of 5 ± 2 µm was laser drilled in each W foil piece (Potomac Photonics Inc.). An optical micrograph of a W support with a hole is shown in Supporting Figure S3. Prior to the graphene transfer, as necessary, some W supports were polished for 30 – 60 s on a diamond lapping film (1 µm average particle size, 3M) to remove sharp edges around the hole resulting from the laser drilling process. Subsequently, all W supports were cleaned by sequential sonication in ethanol for 5 minutes, acetone for 2 minutes, and finally in isopropanol for 2 minutes. To remove organic contaminants on the surface of the foil, the foils were annealed at 1077 °C and 1 bar for 30 minutes under a H2 environment (H2 flow of 70 sccm).

Wet transfer of graphene

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 29

A wet transfer technique was employed to deposit graphene onto the W support.48-49 This involved coating a the graphene/Cu foil with a thin film of polymethyl methacrylate (PMMA) followed by etching the Cu foil, scooping the floating graphene/PMMA film onto the W substrate, annealing the PMMA film, and finally dissolving away the PMMA film. As a result of the CPCVD procedure operating close to the melting temperature of Cu (1083 °C), the graphene/Cu foil bonded to the underlying Cu frame. This helped in maintaining the graphene/Cu foil in a wrinkle-free state during the spin coating of PMMA. Typically, the edges of the Cu frame were masked using adhesive polytetrafluoroethylene (PTFE) tape to prevent coating of the foil’s back side. The PTFE-masked frame was taped to a Si wafer for spin coating. The top surface of graphene/Cu foil was covered with the PMMA solution (MicroChem Corp. 950 PMMA A4, 4% in anisole), and spin coating was carried at 2000 rpm for 2 minutes. The coated film was dried in air on a hot plate at 60 °C for 30 minutes, following which several coupons of the PMMA coated graphene/Cu foil were cut out from the Cu frame. To etch the Cu foil, the PMMA/graphene/Cu coupons (5 mm x 5 mm) were suspended via surface tension on top of a Na2S2O8 bath (Transene Co. Inc.) at room temperature. We found that Na2S2O8 generates fewer detrimental bubbles compared to the commonly used Cu etchant, (NH4)2S2O8, and therefore is advantageous for wrinkle free transfer of graphene. The floating foil was scooped out after 10 minutes and was placed on a fresh sodium persulfate bath to prevent contamination from the graphene released from the backside of the Cu foil. The etching was complete after 60 – 90 minutes. For transfer, a W support and a piece of Si wafer (2 cm x 1 cm) were cleaned in air plasma for 3 minutes (Harrick Plasma PDC-32G) to render their surfaces hydrophilic ensuring a complete spreading of water. The floating graphene/PMMA film was carefully scooped out using the Si wafer, and was placed on the surface of a DI water bath for 5

8 ACS Paragon Plus Environment

Page 9 of 29

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 Journal of Physical Chemistry

minutes. The floating film was rinsed 3 times in fresh water baths to remove the residual etchant. Finally, graphene/PMMA film was scooped onto the W foil support, then dried at 60 °C for 4 hours. To ensure conformal contact between the graphene/PMMA and the W support, the film was annealed in air at 150 °C for 15 minutes, and 200 °C for 10 minutes.50-51 Finally, to remove the PMMA using acetone, the back side of the graphene/PMMAcoated W substrate was masked with a 2 mm thick polydimethylsiloxane (PDMS) piece52 to prevent solvent from entering the cylindrical hole in the support, which could result in capillary force-induced rupture of the graphene film.53 The masked substrate was horizontally submerged in the acetone bath at the room temperature for 5 minutes while holding the PDMS piece on the back mostly out of the acetone bath. The resulting membrane was immersed in isopropanol for 3 seconds to remove acetone-related contamination. Finally, the W foil was dried in air on a hot plate at 60 °C for 4 hours, after which the PDMS piece was removed from the back side of the foil.

Characterization Optical micrographs were acquired using a Carl Zeiss Axio Scope operating in reflection as well as transmission mode. Scanning electron microscopy (SEM) was carried out using a JEOL JSM 6700 microscope operated at 1.5 kV accelerating voltage. No conductive coating was applied on to the samples prior to the SEM imaging. The precise W support hole diameter was determined for each membrane tested from SEM images. The Raman data were collected using a Horiba micro-Raman spectroscopy system (Horiba LabRAM, 532 nm, 2.33 eV, 0.8 mW, 1800 grooves/mm grating, 100x objective). The data acquisition was carried out using LabSpec software (version 5). The spectrometer was

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 29

calibrated using cyclohexane at its most intense peak at 801.8 cm-1. Analysis of the Raman data was done in MATLAB. For calculation of the D and the G peak intensities, the background was subtracted from the Raman data using the least-squares curve fitting tool (lsqnonlin). Ultrahigh vacuum (UHV) scanning tunneling microscopy (STM) was used to identify and characterize the graphene lattice and the structures, spatial distributions, and sizes of the nanopores. An Omicron VT-STM system operating at room temperature and 1.5 x 10-10 mbar base pressure was used to image as-synthesized graphene on the Cu foil after annealing in vacuum to remove surface contaminants. The STM images were processed using the Gwyddion software package.54 Aberration-corrected transmission electron microscopy (ACTEM) was performed using a JEOL 2200MCO with a CEOS image corrector operated at an accelerating voltage of 80 kV. Samples were prepared by transferring the CPCVD graphene onto Si TEM grids with a single square aperture of 0.5 mm x 0.5 mm coated with a thin SiN membrane with an array of 2.5 µm holes (Agar Scientific). Graphene transfer to the grid was carried out using the PMMAmediated wet transfer technique described above. Prior to imaging, the TEM grids were heated in vacuum to desorb surface contaminants.

Pressurization, temperature cycling, and gas permeation testing Evaluation of the pressure and temperature stability as well as gas transport properties of the membranes was carried out in a homemade permeation cell, where the W substrate supporting the graphene membranes was sandwiched between two silicone O-rings. A schematic of the membrane operation is shown in Supporting Figure S4. Typically a pure gas or a mixture of gases at a total pressure of 1.6 – 1.7 bar was fed to the membrane. The feed gases tested

10 ACS Paragon Plus Environment

Page 11 of 29

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 Journal of Physical Chemistry

include H2, He, CH4, CO2, Kr, and SF6. The pressure on the permeate side was maintained at 1.1 bar. Also, the permeate lines were heated to 200 °C to prevent adsorption of the permeated gases in the gas lines. The membrane was housed in a heating mantle (Series O Beaker Mantle, GlassCol) connected to a temperature controller. Membrane temperature was recorded using a thermocouple positioned next to the membrane. The sweep gas (Ar or N2) was connected to a 4-way valve, allowing the module to be operated in two modes. In the continuous mode (valve position: 1 → 3 and 4 → 2, Supporting Figure S4), the sweep gas directed the permeated components to the mass spectrometer (MS, Agilent 5977A coupled with Diablo 5000A real-time gas analyzer). In contrast, in the accumulation mode (valve position: 1 → 2 and 4 → 3, Supporting Figure S4) the sweep gas was allowed to bypass the permeate compartment, thereby allowing the permeating gases to accumulate in the closed loop. The MS was pre-calibrated with respect to the gas composition, yielding a proportional dependence of the MS signal vs. the molar composition (mole %) of gas feed. The MS signal intensities were used to calculate the permeance of each gas species, as described in the Supporting Information Section S1. An example of the MS data collected using the accumulation mode is shown in Supporting Figure S5. The gas permeance lower detection limit was set by the leak rate through the membrane module measured using a tungsten foil with no hole as a control sample. As shown in Supporting Figure S5, control measurements using a bare tungsten support with an open 5 µm hole showed very high gas permeance values (>10-2 mole m-2 s-1 Pa-1) due to the support’s extremely small thickness (50 µm). This large bare support permeance is crucial for the synthesis of high throughput graphene membranes, as otherwise the measured permeance may be limited by the support.42, 55

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 29

Ozone exposure The graphene membranes were exposed to ozone while they were mounted inside the membrane module described above. A pure O2 feed was flowed at 45 sccm through a corona discharge ozone generator (Ozone Engineering LG-7) and onto the feed side of the membrane. The ozone generator power was adjusted to control the ozone concentration following calibration data from the manufacturer.

Results and Discussion We synthesized millimeter-sized graphene crystals (Figure 1A) using controlled-pressure chemical vapor deposition (CPCVD, see Methods).46 Raman characterization (Figure 1B) of graphene transferred to a Si wafer confirmed that the graphene studied herein is a single layer based on the 2D to G peak area ratio, I2D/IG > 2. These spectra also indicate a small density of defects based on the D to G peak area ratio, ID/IG < 0.01.56-57 STM on 20 nm by 20 nm regions of the as-synthesized graphene on the Cu foil revealed the atomic structure of the graphene lattice, as well as some undulations of the surface due to topographic variations in graphene and the underlying Cu surface (Figure 1C). The appearance of the graphene lattice is consistent with STM images of single-layer graphene on Cu reported previously.58-60 Aberration-corrected transmission electron microscopy (ACTEM) images (Figure 1D and Supporting Figure S6) also confirm the pristine atom-thick graphene lattice. The absence of structural defects in these lattice-resolved images of the as-synthesized graphene is consistent with the extremely low density of defects in the graphene film indicated by the Raman spectroscopy.61

12 ACS Paragon Plus Environment

Page 13 of 29

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 Journal of Physical Chemistry

To fabricate the graphene membranes, we used a wet transfer technique48 to suspend graphene over a single 5 ± 2 µm diameter hole in a 50 µm thick W foil membrane support. Using millimeter-sized graphene crystals, we successfully avoided grain boundaries in the suspended graphene domain (Figure 1F). Finally, W-mounted graphene was sealed in a homemade permeation cell using silicone O-rings for characterization of the gas separation properties (Figure 1G). Typically, the feed side (pure gas feed or mixture feed) was pressurized to 1.6 – 1.7 bar, whereas the permeate side connected to a pre-calibrated mass spectrometer (MS) was maintained at 1.1 bar. The temperature of each membrane was varied between 25 - 200 °C. The results of these tests conducted with four different graphene membranes labeled A, B, C, and D, each lasting 17, 36, 7, and 9 days respectively, are shown in Figure 2. For membrane A (Figure 2A), the feed gas was an equimolar mixture of CO2 and CH4. A series of 35 accumulation mode permeance measurements were conducted over 17 days of testing. For all except the final measurement, the permeance of both CH4 and CO2 was at or lower than the detection limit for our measurement technique. These data points are displayed overlaying the dashed horizontal line at the detection limit of 2 × 10-5 mol m-2 s-1 Pa-1. The membrane shows remarkable strength and stability, withstanding aggressive temperature cycling from 25 °C to 200 °C and pressurization for a substantial duration without breaking. Membrane B (Figure 2B) was tested through 53 accumulation mode permeance measurements over 36 days. For these tests, the feed gas composition was changed to include various combinations of H2, He, CH4, CO2, Kr, and SF6, including a six-component mixture feed containing an equal amount of all these gases. As observed for membrane A, there was no detectable permeance of any feed gas through all these measurements, and these data are displayed as points at the detection limit of our technique. These data reveal that the graphene membranes act as effective barriers to the

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 14 of 29

permeation of H2, He, CH4, CO2, Kr, and SF6, with very low or possibly zero flux. The temperature cycling stability in particular is an important result because measuring the temperature dependence of gas permeance through graphene membranes could be an important strategy for extracting activation energies for pore translocation and distinguishing between different permeation mechanisms.18 Eventually, during the final measurement, membrane A partially ruptured, leading to a substantial steady state flux of CO2 and CH4 after switching from the accumulation mode to the continuous detection mode, corresponding to the substantially increased permeance in the final data point in Figure 2A. This confirms that our measurement technique has the ability to detect gas permeation. In the continuous detection mode, membrane rupture can be detected the moment that it occurs with a temporal resolution of approximately 30 s. Membrane C (Figure 2C) was tested via a series of 20 accumulation measurements over 7 days using an equimolar feed of CH4 and CO2. This membrane shows similar results to membrane A, with the permeance of CH4 and CO2 at or below the detection limit for all measurements. To further probe the stability of this membrane, we exposed the graphene to ozone while it was held at 200 °C inside the module. Ozone is a strong oxidant that has the potential to nucleate and grow defects in the membrane through reaction with carbon atoms in the graphene lattice.62-63 Ozone etching may be able to create pores that could be useful for gas separations, and is also a useful probe for investigating the rupture stability of the membranes under oxidizing conditions. Beginning on the 6th day of testing, the membrane was exposed to ozone four times between accumulation measurements for durations of 15 s to 60 s at ozone concentrations of 5 mole % to 15 mole % O3 in O2. After exposure to these large ozone concentrations, the measured permeance of CO2 and CH4 through the membrane remained below

14 ACS Paragon Plus Environment

Page 15 of 29

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 Journal of Physical Chemistry

the detection limit, and the membrane did not rupture. This remarkable chemical and mechanical stability is also a significant result, as this opens the possibility of using O3 exposure as a technique for in situ pore formation or chemical modification without destroying the membrane. Membrane D (Figure 2D) was tested through 26 accumulation measurements over 9 days using an equimolar feed of CH4 and CO2. After confirming that the flux of CH4 and CO2 through this membrane was near or below our detection limit, we exposed this membrane to ozone six times while it was held at 200 °C inside the module. The first several ozone exposures resulted in no change to the permeance through the membrane. Finally, after exposing the membrane to an ozone concentration of 5% mole O3 in O2 for 180 s, we observed increased permeance through the membrane above our detection limit. For the accumulation immediately after this ozone treatment, the measured CH4 permeance was 2.5 × 10-5 mol m-2 s-1 Pa-1, and the CO2 permeance was 2.4 × 10-5 mol m-2 s-1 Pa-1. For the final five accumulation measurements, the measured permeance was increased slightly, with an average CH4 permeance of 6.8 × 10-5 mol m-2 s-1 Pa-1 and an average CO2 permeance of 4.4 × 10-5 mol m-2 s-1 Pa-1, resulting in a CH4/CO2 separation factor of 1.6. The measured permeance did not vary substantially as a function of membrane temperature. These data are consistent with gas permeation through pores in the graphene membrane via the Knudsen direct-gas impingement pathway,13,

18, 64

where the selectivity is

determined by the inverse square root ratio of the molecular weights of the gases, corresponding to an ideal CH4/CO2 separation factor of 1.7. These results provide preliminary evidence that the ozone treatment created pores in the graphene membrane, likely with diameters on the order of ~1 nm to ~100 nm based on the observed Knudsen selectivity.6, 42, 65 Future work will focus on exploring this strategy for perforating the graphene membranes with pores capable of high flux and mixture separation selectivity.

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 16 of 29

The development of new techniques for fabricating sub-nanometer diameter pores in graphene membranes is another critical challenge for enabling selective gas mixture separations. We developed a technique to etch pores into the graphene membrane with the graphene still on the Cu foil via the introduction of O2 into the CVD reactor. We show that in our system hydrogen is not a graphene etchant as previously reported,66 but rather O2 is responsible for graphene etching and pore formation. We began by investigating the defect density in the graphene grown using our normal CPCVD procedure (see Methods) and transferred onto a Si wafer. As shown in Figure 3A, i, Raman mapping measurements confirmed an extremely low ID/IG of (60.0 ± 0.2) × 10-4, corresponding to a large distance between defects, LD, of 152 ± 53 nm (Supporting Information Section S2).67-68 If each of these defects corresponded to one nanopore, this would represent an extremely low pore density of ~47 pores/micron2. However, since it is likely that some defects detected in the Raman spectra do not arise from pores large enough to permit gas translocation, this represents an upper bound estimate on the number of nanopores in this graphene that could be involved in gas separation. ACTEM and STM images of the graphene also show a pristine lattice with very low defect density and no observable nanopores, as discussed previously (Figure 1C, D). To investigate the relationship between O2 exposure and graphene defect density, we studied the Raman ID/IG ratio as a function of the relative concentrations of hydrogen and oxygen in the CPCVD chamber. At the normal graphene growth conditions used here (1077 °C, 108 torr), the residual oxygen concentration in the reactor was approximately 130 ppm, corresponding to a H2:O2 ratio of 7600 (Supporting Information Section S3, Supporting Figures S7 and S8). After the graphene growth steps were completed, we continued exposing the

16 ACS Paragon Plus Environment

Page 17 of 29

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 Journal of Physical Chemistry

graphene to H2:O2 ratio of 7600 at 1077 °C for 3 hours. As a result, the ID/IG ratio increased slightly to (140.0 ± 0.6) × 10-4, corresponding to an LD of 99 ± 28 nm (Figure 3A, ii). Alternatively, a shorter treatment of 15 minutes at a higher O2 concentration (final H2:O2 ratio of 1000) also increased the ID/IG to (150.0 ± 1.1) × 10-4, corresponding to an LD of 96 ± 32 nm, (Figure 3A, iii). We could further increase ID/IG to (43.0 ± 0.3) × 10-4 (LD of 57±15 nm) by etching graphene in a much higher O2 concentration (final H2:O2 ratio of 13) for just 1 minute (Figure 3A, iv). Overall, by increasing the oxygen concentration in the reactor, we could increase the defect density from 47 ± 37 micron-2 to 325 ± 173 micron-2 (Figure 3B). Comparative SEM images (Fig. 3C) of graphene crystals exposed to low and high levels of oxygen (H2:O2 ~ 7600 and 300, respectively) provide additional evidence of the significant role played by oxygen in etching graphene. While the graphene domains do not change in size after 1 hour of exposure to the lower oxygen concentration (H2:O2 ~ 7600), they completely etch away in 1 hour at the higher concentration of oxygen (maximum H2:O2 of 300). These observations support a mechanism whereby residual oxygen is primarily responsible for the intrinsic defects often observed in CVD-grown graphene.42,

69

Furthermore, our observations falsify a competing

mechanistic explanation that H2 is an etchant in the CVD synthesis of graphene under our conditions, and possibly other examples reported in the literature.66, 70-72 While the precise mechanism of this high temperature O2 etching process remains unknown, the occurrence of a reaction between O2 and the carbon atoms in the graphene lattice is consistent with previous studies.63,

73-74

This process likely consists of two steps: pore

nucleation, wherein the first carbon atom is removed from the graphene lattice, and growth, wherein subsequent carbon atoms are extracted at the edges around the initial vacancy, enlarging the nanopore. The kinetics of pore nucleation on pristine graphene are likely much slower than

17 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 18 of 29

pore growth.73 However, previous experimental studies showed that at temperatures above 500 °C, molecular O2 can react with single layer graphene to create defects on the order of 10 – 100 nm in diameter,63, 75 and at temperatures in the range of 260 – 300 °C, sub-nanometer vacancies can be formed.45 While the etching conditions used in this study are substantially different, these prior works suggest that one plausible reaction mechanism consists of O2 dissociative adsorption on the graphene surface followed by reactive desorption of carbon monoxide, yielding the overall reaction C + ½ O2 → CO.73-74 Future work will focus on illuminating the detailed mechanism of this oxidative etching process. Our Raman mapping measurements provide evidence that high temperature O2 etching can increase the defect density in graphene, but these data do not reveal the size or shape of the defects. To investigate whether these defects correspond to nanopores that could be relevant for gas mixture separation, we conducted a thorough STM survey of graphene etched at an H2:O2 ratio of 13. These measurements revealed several pores separated by a distance of approximately 100 – 200 nm (Figure 4). The estimated distance between the pores is in reasonable agreement with that estimated from the amorphization trajectory of graphene etched under these conditions from Raman spectroscopy (LD of 57 ± 15 nm), especially considering the stochastic nature of oxygen etching. The pores (Figure 4, A-E) consist of several missing carbon atoms surrounded by a raised portion at the pore edge and a larger region of lattice distortion extending about 3 – 6 nm into the graphene lattice. The brighter appearance of edges of the pores can be attributed to a combination of distorted irregular bonds that have a higher density of states,76 dangling bonds, interactions with the underlying Cu surface, and heteroatoms bound to the edge.77-78 The extended lattice distortions can be attributed to nanopore induced disruption in the lattice, and have been observed before for point defects.76 The structures observed in the STM images are

18 ACS Paragon Plus Environment

Page 19 of 29

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 Journal of Physical Chemistry

consistent with pores on the order of ~1 nm in diameter, placing these pores in the size regime that is likely of interest for gas mixture separations. Our preliminary efforts to measure gas transport through graphene with these pores did not show permeance above our detection limit. Future studies will focus on using this oxygen etching technique to create a higher density of pores, and studying their performance for mixture separations.

Conclusion In this work, we fabricated graphene gas separation membranes on metal foil supports, and tested them using a custom built membrane module. The membranes show remarkable pressure and temperature stability, as they did not rupture after many days of pressurization and temperature cycling. The permeance measured using an on line mass spectrometer was lower than the detection limit for all membranes, though our data provide a preliminary indication that ozone exposure could create pores in the membranes that permit gas permeation. We also demonstrate a promising high temperature O2 etching technique for creating pores in the graphene. Our platform for graphene membrane fabrication, testing, and perforation provide a pathway for studying the permeance of perforated graphene membranes in future work. This represents important progress towards the development of functional graphene membranes for gas mixture separation applications.

Supporting Information ACTEM images of graphene, schematic of CPCVD reactor, design of Cu frame for CPCVD graphene synthesis, optical micrograph of W support, schematic of gas permeation setup, estimation of defect density from Raman amorphization trajectory, estimation of oxygen concentration in the CPCVD reactor, calculation of gas permeance from mass spectrometer data. 19 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 20 of 29

Acknowledgments This work was supported in part by the U. S. Army Research Laboratory and the U. S. Army Research Office through the Institute for Soldier Nanotechnologies, under contract number W911NF-13-D-0001. Q.H.W. acknowledges support from Arizona State University startup funds.

References 1. Geim, A. K., Graphene: Status and Prospects. Science 2009, 324, 1530-1534. 2. Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A., Electric Field Effect in Atomically Thin Carbon Films. Science 2004, 306, 666-669. 3. Liu, G. P.; Jin, W. Q.; Xu, N. P., Graphene-Based Membranes. Chem. Soc. Rev. 2015, 44, 5016-5030. 4. Yoon, H. W.; Cho, Y. H.; Park, H. B., Graphene-Based Membranes: Status and Prospects. Philos. Trans. R. Soc., A 2016, 374. 5. Yuan, W. J.; Chen, J.; Shi, G. Q., Nanoporous Graphene Materials. Mater. Today 2014, 17, 77-85. 6. Celebi, K.; Buchheim, J.; Wyss, R. M.; Droudian, A.; Gasser, P.; Shorubalko, I.; Kye, J. I.; Lee, C.; Park, H. G., Ultimate Permeation across Atomically Thin Porous Graphene. Science 2014, 344, 289-292. 7. Kim, H. W., et al., Selective Gas Transport through Few-Layered Graphene and Graphene Oxide Membranes. Science 2013, 342, 91-95. 8. Joshi, R. K.; Carbone, P.; Wang, F. C.; Kravets, V. G.; Su, Y.; Grigorieva, I. V.; Wu, H. A.; Geim, A. K.; Nair, R. R., Precise and Ultrafast Molecular Sieving through Graphene Oxide Membranes. Science 2014, 343, 752-754. 9. Li, H.; Song, Z. N.; Zhang, X. J.; Huang, Y.; Li, S. G.; Mao, Y. T.; Ploehn, H. J.; Bao, Y.; Yu, M., Ultrathin, Molecular-Sieving Graphene Oxide Membranes for Selective Hydrogen Separation. Science 2013, 342, 95-98. 10. Koenig, S. P.; Wang, L. D.; Pellegrino, J.; Bunch, J. S., Selective Molecular Sieving through Porous Graphene. Nat. Nanotechnol. 2012, 7, 728-732. 11. Baker, R. W., Future Directions of Membrane Gas Separation Technology. Ind. Eng. Chem. Res. 2002, 41, 1393-1411. 12. Bernardo, P.; Drioli, E.; Golemme, G., Membrane Gas Separation: A Review/State of the Art. Ind. Eng. Chem. Res. 2009, 48, 4638-4663. 13. Koros, W. J.; Fleming, G. K., Membrane-Based Gas Separation. J. Membr. Sci. 1993, 83, 1-80. 14. Sholl, D.; Lively, R., Seven Chemical Separations to Change the World. Nature 2016, 532, 435. 20 ACS Paragon Plus Environment

Page 21 of 29

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 Journal of Physical Chemistry

15. Baker, R. W.; Lokhandwala, K., Natural Gas Processing with Membranes:  An Overview. Ind. Eng. Chem. Res. 2008, 47, 2109-2121. 16. Sun, C. Z.; Wen, B. Y.; Bai, B. F., Application of Nanoporous Graphene Membranes in Natural Gas Processing: Molecular Simulations of CH4/CO2, CH4/H2S and CH4/N2 Separation. Chem. Eng. Sci. 2015, 138, 616-621. 17. Jiang, D. E.; Cooper, V. R.; Dai, S., Porous Graphene as the Ultimate Membrane for Gas Separation. Nano Lett. 2009, 9, 4019-4024. 18. Drahushuk, L. W.; Strano, M. S., Mechanisms of Gas Permeation through Single Layer Graphene Membranes. Langmuir 2012, 28, 16671-16678. 19. Lee, G.-H., et al., High-Strength Chemical-Vapor–Deposited Graphene and Grain Boundaries. Science 2013, 340, 1073-1076. 20. Tsapatsis, M., Toward High-Throughput Zeolite Membranes. Science 2011, 334, 767768. 21. Gemmer, R. Membrane Technology Workshop Summary Report; United State Department of Energy Office of Energy Efficiency & Renewable Energy Advanced Manufacturing Office: 2012. 22. Blankenburg, S.; Bieri, M.; Fasel, R.; Mullen, K.; Pignedoli, C. A.; Passerone, D., Porous Graphene as an Atmospheric Nanofilter. Small 2010, 6, 2266-2271. 23. Li, Y.; Zhou, Z.; Shen, P.; Chen, Z., Two-Dimensional Polyphenylene: Experimentally Available Porous Graphene as a Hydrogen Purification Membrane. Chem. Commun. 2010, 46, 3672-3674. 24. Schrier, J., Helium Separation Using Porous Graphene Membranes. J. Phys. Chem. Lett. 2010, 1, 2284-2287. 25. Du, H. L.; Li, J. Y.; Zhang, J.; Su, G.; Li, X. Y.; Zhao, Y. L., Separation of Hydrogen and Nitrogen Gases with Porous Graphene Membrane. J. Phys. Chem. C 2011, 115, 23261-23266. 26. Wesolowski, R. P.; Terzyk, A. P., Pillared Graphene as a Gas Separation Membrane. Phys. Chem. Chem. Phys. 2011, 13, 17027-17029. 27. Hauser, A. W.; Schwerdtfeger, P., Nanoporous Graphene Membranes for Efficient He3/He-4 Separation. J. Phys. Chem. Lett. 2012, 3, 209-213. 28. Hauser, A. W.; Schwerdtfeger, P., Methane-Selective Nanoporous Graphene Membranes for Gas Purification. Phys. Chem. Chem. Phys. 2012, 14, 13292-13298. 29. Qin, X.; Meng, Q. Y.; Feng, Y. P.; Gao, Y. F., Graphene with Line Defect as a Membrane for Gas Separation: Design Via a First-Principles Modeling. Surf. Sci. 2013, 607, 153-158. 30. Huang, C. X.; Wu, H. P.; Deng, K. M.; Tang, W. H.; Kan, E. J., Improved Permeability and Selectivity in Porous Graphene for Hydrogen Purification. Phys. Chem. Chem. Phys. 2014, 16, 25755-25759. 31. Lei, G. P.; Liu, C.; Xie, H.; Song, F. H., Separation of the Hydrogen Sulfide and Methane Mixture by the Porous Graphene Membrane: Effect of the Charges. Chem. Phys. Lett. 2014, 599, 127-132. 32. Sun, C. Z.; Boutilier, M. S. H.; Au, H.; Poesio, P.; Bai, B. F.; Karnik, R.; Hadjiconstantinou, N. G., Mechanisms of Molecular Permeation through Nanoporous Graphene Membranes. Langmuir 2014, 30, 675-682. 33. Tao, Y. H.; Xue, Q. Z.; Liu, Z. L.; Shan, M. X.; Ling, C. C.; Wu, T. T.; Li, X. F., Tunable Hydrogen Separation in Porous Graphene Membrane: First-Principle and Molecular Dynamic Simulation. ACS Appl. Mater. Interfaces 2014, 6, 8048-8058. 21 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 22 of 29

34. Tsetseris, L.; Pantelides, S. T., Graphene: An Impermeable or Selectively Permeable Membrane for Atomic Species? Carbon 2014, 67, 58-63. 35. Alaghemandi, M., Single Layer Hydrogenated Graphyne Membrane for Selective Hydrogen Separation: A Molecular Dynamics Simulation Study. Chem. Phys. Lett. 2015, 629, 65-69. 36. Darvishi, M.; Foroutan, M., Mechanism of Water Separation from a Gaseous Mixture Via Nanoporous Graphene Using Molecular Dynamics Simulation. RSC Adv. 2015, 5, 8128281294. 37. Liu, H. J.; Chen, Z. F.; Dai, S.; Jiang, D. E., Selectivity Trend of Gas Separation through Nanoporous Graphene. J. Solid State Chem. 2015, 224, 2-6. 38. Nieszporek, K.; Drach, M., Alkane Separation Using Nanoporous Graphene Membranes. Phys. Chem. Chem. Phys. 2015, 17, 1018-1024. 39. Wen, B. Y.; Sun, C. Z.; Bai, B. F., Molecular Dynamics Simulation of the Separation of Ch4/Co2 by Nanoporous Graphene. Acta Physico-Chimica Sinica 2015, 31, 261-267. 40. Wen, B. Y.; Sun, C. Z.; Bai, B. F., Inhibition Effect of a Non-Permeating Component on Gas Permeability of Nanoporous Graphene Membranes. Phys. Chem. Chem. Phys. 2015, 17, 23619-23626. 41. Xu, J.; Sang, P. P.; Xing, W.; Shi, Z. M.; Zhao, L. M.; Guo, W. Y.; Yan, Z. F., Insights into the H-2/Ch4 Separation through Two-Dimensional Graphene Channels: Influence of Edge Functionalization. Nanoscale Res. Lett. 2015, 10. 42. Boutilier, M. S. H.; Sun, C. Z.; O'Hern, S. C.; Au, H.; Hadjiconstantinou, N. G.; Karnik, R., Implications of Permeation through Intrinsic Defects in Graphene on the Design of DefectTolerant Membranes for Gas Separation. ACS Nano 2014, 8, 841-849. 43. Schrier, J., Fluorinated and Nanoporous Graphene Materials as Sorbents for Gas Separations. ACS Appl. Mater. Interfaces 2011, 3, 4451-4458. 44. Liu, H. J.; Dai, S.; Jiang, D. E., Insights into Co2/N-2 Separation through Nanoporous Graphene from Molecular Dynamics. Nanoscale 2013, 5, 9984-9987. 45. Yamada, Y., et al., Subnanometer Vacancy Defects Introduced on Graphene by Oxygen Gas. J. Am. Chem. Soc. 2014, 136, 2232-2235. 46. Yan, Z.; Lin, J.; Peng, Z.; Sun, Z.; Zhu, Y.; Li, L.; Xiang, C.; Samuel, E. L.; Kittrell, C.; Tour, J. M., Toward the Synthesis of Wafer-Scale Single-Crystal Graphene on Copper Foils. ACS Nano 2012, 6, 9110-9117. 47. Yan, Z.; Peng, Z.; Tour, J. M., Chemical Vapor Deposition of Graphene Single Crystals. Acc. Chem. Res. 2014, 47, 1327-1337. 48. Lin, Y.-C.; Jin, C.; Lee, J.-C.; Jen, S.-F.; Suenaga, K.; Chiu, P.-W., Clean Transfer of Graphene for Isolation and Suspension. ACS Nano 2011, 5, 2362-2368. 49. Wang, D.-Y., et al., Clean-Lifting Transfer of Large-Area Residual-Free Graphene Films. Adv. Mater. 2013, 25, 4521-4526. 50. Liang, X., et al., Toward Clean and Crackless Transfer of Graphene. ACS Nano 2011, 5, 9144-9153. 51. Cheng, Z.; Zhou, Q.; Wang, C.; Li, Q.; Wang, C.; Fang, Y., Toward Intrinsic Graphene Surfaces: A Systematic Study on Thermal Annealing and Wet-Chemical Treatment of Sio2Supported Graphene Devices. Nano Lett. 2011, 11, 767-771. 52. Shimizu, S.; Ellison, M.; Aziz, K.; Wang, Q. H.; Ulissi, Z.; Gunther, Z.; Bellisario, D.; Strano, M., Stochastic Pore Blocking and Gating in Pdms–Glass Nanopores from Vapor–Liquid Phase Transitions. J. Phys. Chem. C 2013, 117, 9641-9651. 22 ACS Paragon Plus Environment

Page 23 of 29

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 Journal of Physical Chemistry

53. Lee, C.-K., et al., Monatomic Chemical-Vapor-Deposited Graphene Membranes Bridge a Half-Millimeter-Scale Gap. ACS Nano 2014, 8, 2336-2344. 54. Nečas, D.; Klapetek, P., Gwyddion: An Open-Source Software for Spm Data Analysis. Cent. Eur. J. Phys. 2012, 10, 181-188. 55. Agrawal, K. V., et al., Oriented Mfi Membranes by Gel-Less Secondary Growth of Sub100 Nm Mfi-Nanosheet Seed Layers. Adv. Mater. 2015, 27, 3243-3249. 56. Ferrari, A. C., Raman Spectroscopy of Graphene and Graphite: Disorder, ElectronPhonon Coupling, Doping and Nonadiabatic Effects. Solid State Commun. 2007, 143, 47-57. 57. Ferrari, A. C.; Basko, D. M., Raman Spectroscopy as a Versatile Tool for Studying the Properties of Graphene. Nat. Nanotechnol. 2013, 8, 235-246. 58. Gao, L.; Guest, J. R.; Guisinger, N. P., Epitaxial Graphene on Cu(111). Nano Lett. 2010, 10, 3512-3516. 59. Yu, Q., et al., Control and Characterization of Individual Grains and Grain Boundaries in Graphene Grown by Chemical Vapour Deposition. Nat. Mater. 2011, 10, 443-449. 60. Zhao, L.; Rim, K. T.; Zhou, H.; He, R.; Heinz, T. F.; Pinczuk, A.; Flynn, G. W.; Pasupathy, A. N., Influence of Copper Crystal Surface on the Cvd Growth of Large Area Monolayer Graphene. Solid State Commun. 2011, 151, 509-513. 61. Robertson, A. W.; Lee, G.-D.; He, K.; Gong, C.; Chen, Q.; Yoon, E.; Kirkland, A. I.; Warner, J. H., Atomic Structure of Graphene Subnanometer Pores. ACS Nano 2015, 9, 1159911607. 62. Lee, G.; Lee, B.; Kim, J.; Cho, K., Ozone Adsorption on Graphene: Ab Initio Study and Experimental Validation. J. Phys. Chem. C 2009, 113, 14225-14229. 63. Liu, L.; Ryu, S.; Tomasik, M. R.; Stolyarova, E.; Jung, N.; Hybertsen, M. S.; Steigerwald, M. L.; Brus, L. E.; Flynn, G. W., Graphene Oxidation: Thickness-Dependent Etching and Strong Chemical Doping. Nano Lett. 2008, 8, 1965-1970. 64. Knudsen, M., Die Gesetze Der Molekularströmung Und Der Inneren Reibungsströmung Der Gase Durch Röhren. Ann. Phys. (Berlin, Ger.) 1909, 333, 75-130. 65. Jennings, S. G., The Mean Free Path in Air. J. Aerosol Sci. 1988, 19, 159-166. 66. Vlassiouk, I.; Regmi, M.; Fulvio, P.; Dai, S.; Datskos, P.; Eres, G.; Smirnov, S., Role of Hydrogen in Chemical Vapor Deposition Growth of Large Single-Crystal Graphene. ACS Nano 2011, 5, 6069-6076. 67. Cançado, L. G.; Jorio, A.; Ferreira, E. H. M.; Stavale, F.; Achete, C. A.; Capaz, R. B.; Moutinho, M. V. O.; Lombardo, A.; Kulmala, T. S.; Ferrari, A. C., Quantifying Defects in Graphene Via Raman Spectroscopy at Different Excitation Energies. Nano Lett. 2011, 11, 31903196. 68. Lucchese, M. M.; Stavale, F.; Ferreira, E. H. M.; Vilani, C.; Moutinho, M. V. O.; Capaz, R. B.; Achete, C. A.; Jorio, A., Quantifying Ion-Induced Defects and Raman Relaxation Length in Graphene. Carbon 2010, 48, 1592-1597. 69. Banhart, F.; Kotakoski, J.; Krasheninnikov, A. V., Structural Defects in Graphene. ACS Nano 2011, 5, 26-41. 70. Choubak, S.; Levesque, P. L.; Gaufres, E.; Biron, M.; Desjardins, P.; Martel, R., Graphene Cvd: Interplay between Growth and Etching on Morphology and Stacking by Hydrogen and Oxidizing Impurities. J. Phys. Chem. C 2014, 118, 21532-21540. 71. Zhang, Y.; Li, Z.; Kim, P.; Zhang, L.; Zhou, C., Anisotropic Hydrogen Etching of Chemical Vapor Deposited Graphene. ACS Nano 2012, 6, 126-132.

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 24 of 29

72. Choubak, S.; Biron, M.; Levesque, P. L.; Martel, R.; Desjardins, P., No Graphene Etching in Purified Hydrogen. J. Phys. Chem. Lett. 2013, 4, 1100-1103. 73. Frankcombe, T. J.; Bhatia, S. K.; Smith, S. C., Ab Initio Modelling of Basal Plane Oxidation of Graphenes and Implications for Modelling Char Combustion. Carbon 2002, 40, 2341-2349. 74. Carlsson, J. M.; Hanke, F.; Linic, S.; Scheffler, M., Two-Step Mechanism for LowTemperature Oxidation of Vacancies in Graphene. Phys. Rev. Lett. 2009, 102, 166104. 75. Jones, J. D.; Morris, C. F.; Verbeck, G. F.; Perez, J. M., Oxidative Pit Formation in Pristine, Hydrogenated and Dehydrogenated Graphene. Appl. Surf. Sci. 2013, 264, 853-863. 76. Ugeda, M. M.; Fernández-Torre, D.; Brihuega, I.; Pou, P.; Martínez-Galera, A. J.; Pérez, R.; Gómez-Rodríguez, J. M., Point Defects on Graphene on Metals. Phys. Rev. Lett. 2011, 107, 116803. 77. Kano, E.; Hashimoto, A.; Kaneko, T.; Tajima, N.; Ohno, T.; Takeguchi, M., Interactions between C and Cu Atoms in Single-Layer Graphene: Direct Observation and Modelling. Nanoscale 2016, 8, 529-535. 78. Lee, J.; Zhou, W.; Pennycook, S. J.; Idrobo, J.-C.; Pantelides, S. T., Direct Visualization of Reversible Dynamics in a Si6 Cluster Embedded in a Graphene Pore. Nat. Commun. 2013, 4, 1650.

24 ACS Paragon Plus Environment

Page 25 of 29

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 Journal of Physical Chemistry

Figure 1. Synthesis and assembly of graphene membranes. A) Millimeter-size graphene single crystals on a Cu foil synthesized by the CPCVD technique. B) Raman spectrum of graphene transferred to Si, and a histogram of I2D/IG from Raman mapping (inset). C) STM image of the surface of as-synthesized graphene on Cu showing atomic lattice (bias voltage = -0.1V, tunneling current = 0.5 nA). D) ACTEM image of suspended graphene. A few patches on the surface are PMMA residues from the wet transfer. E) Graphene transferred onto W substrate with a single hole (highlighted by the circle). The dashed lines are visual guides for grain boundaries. F) An enlarged image of the suspended graphene membrane on the W substrate. G) A schematic showing (i) the W substrate, (ii) transfer of the CPCVD graphene onto the W substrate, and then (iii) assembly of the supported membrane inside a homemade membrane module. Arrows indicate the direction of gas flow on the feed and the permeate sides.

25 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 26 of 29

Figure 2. Results of pressurization, temperature cycling, and gas permeation measurements for four graphene membranes A, B, C, and D. Permeance data points overlaying the detection limit of 2 × 10-5 mol m-2 s-1 Pa-1 indicate that the measured permeance was at or below this value. The approximate permeance of a bare support with no graphene is indicated by the dashed line near the top of each panel. Gray stars denote ozone exposures carried out with the graphene mounted inside the membrane module between accumulation mode permeance measurements. The ozone concentrations refer to the mole % of O3 with the balance O2.

26 ACS Paragon Plus Environment

Page 27 of 29

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 Journal of Physical Chemistry

Figure 3. Characterization of nanopores in the CPCVD graphene. A) Histograms of the ID/IG ratio for CPCVD graphene exposed to different concentration of oxygen at the synthesis temperature. B) Defect density in graphene as a function of maximum oxygen concentration in the CPCVD reactor. C) SEM images of graphene single crystals before and after etching with various concentrations of oxygen. Graphene crystals completely etch away in an hour when the H2:O2 ratio was decreased to 300.

27 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 28 of 29

Figure 4. STM images of the nanopores in the as-synthesized CPCVD graphene on a Cu foil subjected to 1 min etching at a H2:O2 ratio of up to 13. The bias voltage was -0.1V. Tunneling currents for (A) and (B) were 0.4 nA, and for (C), (D), and (E) were 0.5 nA.

28 ACS Paragon Plus Environment

Page 29 of 29

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 Journal of Physical Chemistry

TOC Graphic

29 ACS Paragon Plus Environment