Thermal release of hydrogen retained in multilayer graphene films prepared by mist-chemical vapor deposition

    Thermal release of hydrogen retained in multilayer graphene films prepared by mist-chemical vapor deposition B. Tsuchiya, N. Matsunami, S. Bandow, S. Nagata PII: DOI: Reference:

S0925-9635(15)30098-4 doi: 10.1016/j.diamond.2015.11.019 DIAMAT 6521

To appear in:

Diamond & Related Materials

Received date: Revised date: Accepted date:

24 August 2015 14 November 2015 27 November 2015

Please cite this article as: B. Tsuchiya, N. Matsunami, S. Bandow, S. Nagata, Thermal release of hydrogen retained in multilayer graphene films prepared by mist-chemical vapor deposition, Diamond & Related Materials (2015), doi: 10.1016/j.diamond.2015.11.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

T

Thermal release of hydrogen retained in multilayer graphene films prepared by mist-chemical vapor deposition

IP

B. Tsuchiya1,*, N. Matsunami1, S. Bandow2, S. Nagata3

Department of General Education, Faculty of Science and Technology, Meijo University, 1-501, Shiogamaguchi, Tempaku-ku, Nagoya 468-8502, Japan

2

Department of Applied Chemistry, Faculty of Science and Technology, Meijo University, 1-501, Shiogamaguchi, Tempaku-ku, Nagoya 468-8502, Japan Institute for Materials Research, Tohoku University, 2-1-1, Katahira, Aoba-ku, Sendai 980-8577, Japan

MA

3

NU

SC R

1

*Corresponding author: E-mail: [email protected], Tel: +81-52-832-1151, Fax: +81-52-832-1179

ED

Abstract

In this study, we investigated the absorption and thermal desorption processes of H and

PT

H2O and the thickness of multilayer graphene films deposited on Cu foils using a mist-

AC CE

chemical vapor deposition method. Ion beam analysis techniques such as nuclear reaction analysis (NRA), elastic recoil detection (ERD), and Rutherford backscattering spectrometry (RBS) were employed. The RBS measurements revealed that the thickness of the multilayer graphene films was approximately 8±3 nm (24±9 layers). The depth distribution of H was analyzed using NRA and ERD. Based on these measurements, the residual H/C ratio for multilayer graphene was estimated to be approximately 0.03 in the bulk and 0.88 on the topmost surface. Additionally, the thermal desorption temperature for H from the multilayer graphene film was less than 373 K, which was much lower than that from isotropic graphite bulk (approximately 673 K). These results suggest that the thermal release of H did not occur because of desorption from sp2- and sp3-hybridized C atoms, such as intercalation and defect sites. Instead, it occurred owing to the desorption of H2O adsorbed near the surface.

ACCEPTED MANUSCRIPT

T

Keywords: graphene, hydrogen, water, absorption, desorption, nuclear reaction analysis, elastic recoil detection, Rutherford backscattering spectrometry

IP

Introduction

SC R

Graphene, which is composed of a monolayer of two-dimensional honeycomb lattice of sp2-hybridized orbitals or a single sheet of graphite, is expected to have potential applications as a H source for H-based fuel cells [1-5]. The ratio of residual H intercalated in

NU

the layers and defects to C (H/C) has been reported to be approximately 0.4 for isotropic

MA

bulk graphite because of this material’s C deficiency [6]. In a recent study, the H uptakes by graphene measured at 77 and 296 K were 5.5 wt.% (30 bar) and 0.89 wt.% (120 bar),

ED

respectively [4, 5]. However, some types of H-trapping sites in graphene have not been experimentally investigated. It is especially necessary to investigate the depth distribution of

H storage.

PT

H and its adsorption to and desorption from graphene to utilize multilayer graphene films for

AC CE

Therefore, the aim of this study is to investigate the thickness of multilayer graphene films produced by mist-chemical vapor deposition (CVD), the depth distribution of H in multilayer graphene films, and the characteristics of H and H2O adsorption to and desorption from multilayer graphene films after heating to 573 K under vacuum. For this purpose, techniques including Rutherford backscattering spectrometry (RBS), nuclear reaction analysis (NRA), and elastic recoil detection (ERD) were used.

2. Experimental methods Multilayer graphene films were fabricated by the CVD method using an ultrasonically generated methanol mist. After the reaction vessel was evacuated to below 1.3 × 10−2 Pa and heated to approximately 1198 K in a furnace, Ar gas containing 3% H2 and

ACCEPTED MANUSCRIPT

the methanol mist were introduced into the vessel at a flow rate of approximately 1.3 L/min. Subsequently, a Cu foil (dimensions: 30 × 15 × 1.0 mm3) was quickly mounted onto the

IP

T

vessel for 2 min, and graphene films were deposited on the Cu foil.

To measure the thickness of the prepared multilayer graphene films, the Cu foils

SC R

were chemically etched using CuSO4 and FeCl3 solutions, and then, the films were transferred onto SiO2-glass substrates [7]. The thickness of the fabricated multilayer

NU

graphene film was evaluated using the RBS technique with a 1.8-MeV He+ ion probe beam produced from the Van de Graaff accelerator at Nagoya University. The multilayer graphene

MA

film was irradiated with He+ ions at an incident angle of 90° with respect to the film surface at room temperature. The back-scattered He+ ions generated by elastic collisions with C, O,

ED

Si, and Cu atoms were detected at an angle of 170° with respect to the incident He+ ion beam using a solid state detector (SSD) [8]. The thickness of the film or its number of layers

PT

was estimated from the energy of the back-scattered He+ ions generated by elastic collisions

AC CE

with C. The estimated number of layers was compared with that obtained from optical transmission measurements. To estimate the H distribution in the multilayer graphene films, NRA and ERD were conducted with 6.38–6.50-MeV N+ ions and 2.8-MeV He2+ ions, respectively, generated by tandem accelerators installed at the Japan Atomic Energy Agency and the Institute for Materials Research of Tohoku University. For the NRA measurements, the direction of the incident N+ ions was normal to the surface of the multilayer graphene films. Subsequently, 4.44-MeV gamma rays emitted during the nuclear resonance reaction of N+ ions with H atoms, H(15N, )12C, were observed using a NaI (Tl) scintillation detector connected to a 1024-channel multichannel analyzer [9]. For the ERD measurements, the He2+ ions were incident at an angle of 15° with respect to the sample surface. The forward-recoiled H+ ions

ACCEPTED MANUSCRIPT generated from collisions with He2+ ions were then detected at a scattering angle of 30° with respect to the incident He2+ ion beam using an SSD equipped with an absorber made of a

IP

T

12.0-µm-thick Al film [10].

SC R

3. Experimental results and discussion

Fig. 1 shows typical RBS spectra of multilayer graphene-coated SiO2

NU

(graphene/SiO2) and SiO2 without the graphene coating. He+ ion beams with an energy of 1.8-MeV were used for these measurements in addition to an expanded beam in the channel

MA

number range of 170–250. The small peak observed in the channel number range of 205– 220 is attributed to C. The thickness of the multilayer graphene films was determined to be

ED

approximately 8±3 nm (24±9 layers) by subtracting the O and Si background yields of the SiO2 substrate from the total C yield. The Rutherford backscattering cross-sections of He+

PT

ions for the constituent elements of graphene/SiO2, the stopping cross-sections for He+ ions,

AC CE

and the solid angle of detection were also considered [11]. The density of the multilayer graphene films and the interlayer distance c between two graphite layers were assumed to be 2.2 g/cm3 (11 × 1022 C/cm3) and 0.336 nm [12], respectively. Fig. 2 shows the number of stacked layers of graphene films derived from the RBS and optical transmission measurements. The optical transmission intensity was assumed to decrease by 2.3% per graphene layer at the optical wavelength of 550 nm [13]. The thickness of the multilayer graphene films derived from the RBS measurements was approximately 1.5 times higher than that determined from the optical transmission measurements (optical thickness): approximately 17±4 layers (approximately 6±1 nm). This discrepancy is attributable to the accuracy of the value used for the He+ ion stopping power and the scattered yield of the small peak. In addition, the RBS spectra revealed that some

ACCEPTED MANUSCRIPT

impurities such as Cu, Fe, and S were completely removed from the multilayer graphene films by thoroughly rinsing after etching the Cu foils.

IP

T

Figs. 3(a) and (b) show the typical NRA spectra of emitted gamma rays from -phase (face-centered tetragonal structure) non-stoichiometric zirconium hydride (ZrH1.99) and

SC R

multilayer graphene-coated Cu (graphene/Cu), respectively. The measurements were conducted at room temperature under vacuum using 6.38–6.50-MeV N+ ions. The horizontal

NU

axis in the NRA spectra represents the incident energy (MeV) of the N+ ions, which indicates the ions’ depth of penetration from the top-most surface, whereas the vertical axis

MA

represents the peak counts in the gamma ray spectrum per unit of the N+ ionic current, which reflects the concentration of H retained in the graphene/Cu. Fig. 3(a) shows that the H

ED

concentration at a depth of approximately 70 nm from the top-most surface of ZrH1.99 was much lower than that in the bulk because of the presence of zirconium oxides on the surface.

PT

The depth profile of H retained in graphene/Cu can be determined from the NRA spectrum

AC CE

shown in Fig. 3(b). The H concentrations on the surface and in the bulk of the multilayer graphene films (NH on graphene and NH in graphene) were determined to be approximately 9.63 × 1022 and 3.48 × 1021 H/cm3, respectively, based on the H concentration in ZrH1.99 (approximately 8.58×1022 H/cm3), as shown in Fig. 3(a). In other words, the residual H/C ratio in the multilayer graphene films was estimated to be approximately 0.03 in the bulk and 0.88 on the surface, considering the density of the multilayer graphene films (11 × 1022 C/cm3). Based on the bulk result, we expect that a single H atom occupies only approximately five six-membered graphene rings. The high H concentration on the surface may be attributed to adsorbed H2O, suggesting the excellent hydrophilicity of the surface. Furthermore, Fig. 3(b) indicates that although H was adsorbed at the graphene/Cu interface,

ACCEPTED MANUSCRIPT

no H absorption occurred in the bulk of the Cu foil. The H concentration at the graphene/Cu interface (NH at graphene/Cu) was determined to be approximately 2.19 × 1021 H/cm3.

IP

T

Figs. 4(a) and (b) present typical ERD spectra of recoiled H+ ions from Cu and graphene/Cu, respectively, after isochronal annealing for 10 min at several temperatures up

SC R

to 573 K. The spectra were measured at room temperature under vacuum using 2.8-MeV He2+ ions. The horizontal axes (channel number) in the ERD spectra correspond to the

NU

energies of the recoiled H+ ions and the depth from the top-most surface, which is approximately 8 nm/channel. The depth detection limit was approximately 600 nm. The

MA

vertical axes (counts per channel) in the ERD spectra reflect the area densities of H retained in Cu and graphene/Cu. Small and large amounts of H were observed in the Cu and the as-

ED

prepared multilayer graphene-coated Cu foils, respectively, as shown in Figs. 4(a) and (b). The H/C ratio in the as-prepared multilayer graphene film was estimated to be

PT

approximately 0.74 and was determined by considering the He+ ion fluence, the highest peak

AC CE

height, the elastic collision cross-sections of He+ ions with H, the stopping cross-sections of He+ ions for C and H, and the solid angle of detection [11]. The obtained value is similar to the H content at the surface determined by NRA. Because the depth resolution exceeds approximately 100 nm, it is not possible to distinguish between the H content at the surface and that in the bulk. In addition, the formation of bonds between H atoms, H2 molecules, – OH, and H2O cannot be identified. However, it can be inferred that the ERD spectra in Figs. 4(a) and (b) reveal the H depth distributions caused by H2O adsorbed on the Cu foil and H2O and -OH at the surface and H in the bulk of graphene/Cu, respectively. The residual H concentration retained in graphene/Cu significantly decreased as the annealing temperature increased to 573 K, whereas that in Cu decreased slightly. Fig. 4(b) also indicates that H desorption occurred uniformly at all depths in the multilayer graphene film.

ACCEPTED MANUSCRIPT

Fig. 5 shows the changes in the residual H concentration in the multilayer graphene film after isothermal annealing for 10 min at 373 and 573 K under vacuum. These changes

IP

T

in H concentration are compared with those in isotropic bulk graphite. The fraction of the retained H along the vertical axis was obtained by normalizing the total integral counts of

SC R

the ERD spectra collected after annealing at 373 and 573 K to that for the as-prepared film. The temperature corresponding to the thermal desorption of H from the multilayer graphene

NU

film was less than 373 K, which was much lower than that (approximately 673 K) for isotropic graphite bulk. In addition, the H release rate for the multilayer graphene film was

MA

lower than that for bulk graphite. This result indicates that the H desorption process at some trapping sites may be different from that in the graphite bulk, including trapping sites with

ED

sp2- and sp3-hybridized orbitals, such as intercalation sites between layers and substitutional sites, because of this material’s C deficiency. Therefore, based on the NRA and ERD results,

PT

it can be concluded that the multilayer graphene films possess excellent hydrophilic

AC CE

characteristics and that significant amounts of H2O adsorbed onto the top-most surface are released by heating to temperatures below 373 K.

4. Summary

The residual H absorbed in the approximately 8±3 nm (24±9 layers) multilayer graphene films coated on SiO2 and Cu was measured using ion beam analysis techniques such as NRA, RBS, and ERD with N+, 1.8-MeV He+, and 2.8-MeV He2+ ion probe beams, respectively, produced by Van de Graaff and tandem accelerators. From the H depth distribution results measured by NRA and ERD (to a depth of approximately 600 nm), the residual H/C ratio in the bulk of the multilayer graphene films was determined to be approximately 0.03, which was much lower than that (approximately 0.4) in the isotropic graphite bulk. However, a

ACCEPTED MANUSCRIPT significant amount of H (9.63 × 1022 H/cm3) was absorbed on the top-most surface of the multilayer graphene films. In the case of the multilayer graphene film, the thermal

IP

T

desorption temperature of H was less than 373 K, which was much lower than that of the isotropic graphite bulk: approximately 673 K. Further, at a given depth, the H concentration

SC R

in the multilayer graphene film decreased as the annealing temperature increased in the range of 298–573 K. These results show that the thermal release of H does not occur via

NU

desorption from C atoms with sp2- and sp3-hybridized orbitals, such as intercalation and

MA

defects. Instead, it occurs through the desorption of H2O adsorbed near the surface.

References

AC CE

PT

ED

[1] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films, Science 306 (2004) 666-669. [2] K. Spyrou, D. Gournis, P. Rudolf, Hydrogen storage in graphene-based materials efforts towards enhanced hydrogen absorption, ECS J. Solid State Sci. and Technol. 2 (2013) M3160-M3169. [3] V. Tozzini, V. Pellegrini, Prospects for hydrogen storage in graphene, Phys. Chem. Chem. Phys. 15 (2013) 80-89. [4] A.G. Klechikov, G. Mercier, P. Merino, S. Blanco, C. Merino, A.V. Talyzin, Hydrogen storage in bulk graphene-related materials, Microporous and Mesoporous Materials 210 (2015) 46-51. [5] I.A. Baburin, G. Seifert, A. Klechikov, A. Talyzin, G. Mercier, Hydrogen adsorption by perforated graphene, Int. J. Hydrogen Energy 40 (2015) 6594-6599. [6] B. Tsuchiya, S. Nagata, T. Shikama, Oxygen ion-induced detrapping of hydrogen retained in graphite, Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 426-430. [7] T. Mizuno, M. Takizawa, B. Tsuchiya, M. Jinno, S. Bandow, A nitrogen-doped graphene film prepared by chemical vapor deposition of a methanol mist containing methylated melamine resine, Appl. Phys. A 113 (2013) 645-650. [8] N. Shindei, N. Matsunami, O. Fukuoka, M. Tazawa, T. Shimura, Y. Chimi, M. Sataka, Reaction of implanted N isotope with SiO2 near Si3N4-film and SiO2-substrate interface, J. Nucl. Sci. and Technol. 43 (2006) 382-385. [9] R.A. Livingston, J.S. Schweitzer, C. Rolfsc, H.-W. Becker, S. Kubsky, T. Spillane, J. Zickefoose, M. Castellote, P.G. de Viedma, J. Cheung, Heavy ion beam measurement of the hydration of cementitious materials, Applied Radiation and Isotopes 68 (2010) 683687. [10] B. Tsuchiya, K. Morita, S. Nagata, K. Toh, T. Shikama, Temperature dependence of uptake and release of H in near surfaces, Nucl. Instr. and Meth. in Phys. Res. B 257 (2007)

ACCEPTED MANUSCRIPT

SC R

IP

T

541-544. [11] J. F. Ziegler, J. P. Biersack, U. Littmark, The Stopping and Range of Ions in Solids, Pergamon Press, New York (1985). [12] M. Birowska, K. Milowska, J.A. Majewski, Van Der Waals Density Functionals for Graphene Layers and Graphite, Acta Physica Polonica A 120 (2011) 845-848. [13] R.R. Nair, P. Blake, A.N. Grigorenko, K.S. Novoselov, T.J. Booth, T. Stauber, N.M.R. Peres, A.K. Geim, Fine structure constant defines visual transparency of graphene, Science 320 (2008) 1308-1314.

Figure Captions

Typical RBS spectra of back-scattered He+ ions obtained from multilayer graphene-

NU

Fig. 1

Comparison of the number of stacked multilayer graphene film layers determined

ED

Fig. 2

MA

coated SiO2 (graphene/SiO2) and SiO2 measured using 1.8-MeV He+ ion probe beams.

Fig. 3

PT

from RBS and optical transmission measurements at 550 nm.

Typical NRA spectra of (a) zirconium hydride (ZrH1.99) and (b) multilayer

AC CE

graphene-coated Cu (graphene/Cu) measured using N+ ion probe beams with incident energies above 6.38-MeV.

Fig. 4

Typical ERD spectra of recoiled H+ ions from (a) Cu foil and (b) multilayer

graphene-coated Cu (graphene/Cu) after isochronal annealing for 10 min at various temperatures up to 573 K. The spectra were collected using 2.8-MeV He2+ ion probe beams.

Fig. 5

Dependence of the fraction of H retained in multilayer graphene-coated Cu

(graphene/Cu) on temperature (up to 573 K). The fraction of H was determined based on the ERD measurements and compared with the value obtained in the isotropic graphite bulk.

ACCEPTED MANUSCRIPT

1.8-MeV He + graphene/SiO 2

RBS spectra/90°

SiO2

IP

Counts

Counts

C

T

C

SC R

O

200

0

100

MA

NU

Si

250

Channel Number

200

300

400

500

600

700

800

30

35

Channel Number

ED

Fig.1(a) B.Tsuchiya et al.

PT

30

AC CE

Number of Stacked Layers by RBS Measurements

35

25 20 15 10

5 0

0

5

10

15

20

25

Number of Stacked Layers by Optical Transmission Measurements Fig.2 B.Tsuchiya et al.

ACCEPTED MANUSCRIPT

Depth (nm) 0

100

200

300

400

500

600

700

(a) ZrH 1.99

T

11000

IP

10000 9000

SC R

8000 7000 6000 5000

22 NZrH1.99 H/cm 3 ZrH =8.58x10

NU

4000

1.99

3000

MA

2000 1000 0 6.40

6.60

6.80

7.00

7.20

7.40

ED

N+ Ion Incident Energy (MeV)

PT

Fig.3(a) B.Tsuchiya et al.

AC CE

Gamma Ray Yield (Counts/ C)

12000

7.60

ACCEPTED MANUSCRIPT

Depth (nm) 350

0

10

20

30

40

50

60

70

on graphene

T

NH

=9.63x10 22 H/cm 3

IP

300

SC R

250 200 150 100

=3.48x10 21 H/cm 3

NU

NH

in graphene

NH

50

MA

Gamma Ray Yield (Counts/ C)

(b) graphene/Cu

0 6.40

6.42

ED

6.38

6.44

at graphene/Cu interface 21 3

=2.19x10 6.46

H/cm

6.48

N+ Ion Incident Energy (MeV)

AC CE

PT

Fig.3(b) B.Tsuchiya et al.

6.50

ACCEPTED MANUSCRIPT

40 (a) Cu, isochronal annealing for 10 min 35 as-prepared 373 K 428 K 483 K 573 K

T

IP

25

SC R

Counts

30

20

NU

15 10

0

0

MA

5 50

100

150

200

250

ED

Channel Number

40

(b) graphene/Cu, isochronal annealing for 10 min

AC CE

35

500

Counts

30

[nm] 300

100 0

as-prepared 373 K 573 K

25

1.0 0.9 0.8 0.7 0.6

20

0.5

15

0.4 0.3

10

0.2

5

0.1

0

0.0 250

0

50

100

150

Channel Number Fig.4(b) B.Tsuchiya et al.

200

H/C

PT

Fig.4(a) B.Tsuchiya et al.

ACCEPTED MANUSCRIPT

graphite graphene/Cu

T

1.0

SC R

IP

0.8 0.6

NU

0.4 0.2 0.0

300

400

500

MA

Fraction of Retained H

1.2

600

700

800

Temperature

AC CE

PT

ED

Fig.5 B.Tsuchiya et al.

900

(K)

1000 1100 1200

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

SC R

IP

T

We declare as follows; In this present study, the hydrogen (H)- and water (H2O)absorption and thermal desorption as well as graphene thickness measurements on and into graphene films, were in situ investigated using ion beam analyses such as nuclear reaction analysis (NRA), elastic recoil detection (ERD), and Rutherford backscattering spectrometry (RBS). In particular, the depth profile of H on and in the graphene as well as the thermal behavior of H and H2O on the graphene films were clearly observed. 1) the material has not been published in whole or in part elsewhere; 2) the paper is not currently being considered for publication elsewhere; 3) all authors have been personally and actively involved in substantive work leading to the paper, and will hold themselves jointly and individually responsible for its content.

ACCEPTED MANUSCRIPT

Graphical Abstract



IP

T

The title of the manuscript: “Thermal release of hydrogen retained in multilayer graphene films prepared by mist-

1.8-MeV He + graphene/SiO 2

RBS spectra/90°

SiO2

Counts

Counts

C O

200

200

300

400

500

600

ED

100

PT

Channel Number

AC CE

0

250

Channel Number

MA

Si

0

300

Depth (nm)

10

20

30

40

700

800

50

60

70

NRA spectrum, graphene/Cu NH

on graphene

=9.63x10 22 H/cm 3

NU

C

350

Gamma Ray Yield (Counts/ C)

Thickness of Multilayer Graphene Films

SC R

chemical vapor deposition”

250 Hydrogen Depth Profile

200 150 NH

100

in graphene

=3.48x10 21 H/cm 3 NH

50

at graphene/Cu interface 21 3

=2.19x10

H/cm

0 6.38

6.40

6.42

6.44

6.46

6.48

N+ Ion Incident Energy (MeV)

6.50

ACCEPTED MANUSCRIPT

Highlights

T

IP

SC R

NU MA ED



PT



Multilayer graphene films were fabricated by a chemical vapor deposition (CVD) method using an ultrasonically generated methanol mist. Rutherford backscattering spectrometry (RBS) measurements revealed that the thickness of the multilayer graphene films was approximately 8±3 nm (24±9 layers). The absorption and thermal desorption processes of H and H2O absorbed in the multilayer graphene films were investigated using nuclear reaction analysis (NRA) and elastic recoil detection (ERD) measurements.

AC CE

