- Email: [email protected]
Investigation of the magnetic properties of proton irradiated type Ib HPHT diamond
Diamond & Related Materials 64 (2016) 197–201
Contents lists available at ScienceDirect
Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond
Investigation of the magnetic properties of proton irradiated type Ib HPHT diamond N. Daya a,e,⁎, E. Sideras-Haddad a,e, T.N. Makgato a,e, M. García-Hernández b, A. Climent-Font c,d, A. Zucchiatti c, M.A. Ramos c,d a
School of Physics, University of the Witwatersrand, Johannesburg 2050, South Africa Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, E-28049 Madrid, Spain Centro de Microanálisis de Materiales (CMAM), Universidad Autónoma de Madrid, Cantoblanco, E-28049 Madrid, Spain d Instituto de Ciencia de Materiales “Nicolás Cabrera”, Universidad Autónoma de Madrid, Cantoblanco, E-28049 Madrid, Spain e Centre of Excellence in Strong Materials, Physics Building, University of the Witwatersrand, Johannesburg 2050, South Africa b c
a r t i c l e
i n f o
Article history: Received 1 November 2015 Received in revised form 22 February 2016 Accepted 23 February 2016 Available online 26 February 2016 Keywords: Amorphous carbon Diamond Magnetism
a b s t r a c t Nitrogen rich type Ib HPHT grown synthetic diamonds were investigated following large area irradiation of the samples using 2.2 MeV protons. Studies on possible magnetic properties induced after the irradiation were performed using a SQUID magnetometer. Magnetisation measurements of pristine (unirradiated) control samples revealed a superparamagnetic-like signal at 300 K. After the proton irradiation, a Curie-like paramagnetic curve was observed for thermal cycles at an applied magnetic field of 2 kOe which exhibits a transition at temperatures around 50–55 K, with hysteretic behaviour below these temperatures. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The magnetic properties of carbon have gained significant attention over the recent years with extensive experimental and theoretical research continuously been performed on this challenging topic. Magnetic ordering in this class of carbon-based materials at room temperatures could lead to revolutionary new technologies in nanotechnology, biomedicine and many other fields [1] since they are light weight and inexpensive to produce [2]. Such materials present unique prospects in terms of spin-based electronics due to the fact that the weak spin–orbit coupling in carbon results in the critically desired long diffusion lengths and coherence times. However, ferromagnetism in carbon allotropes is rather unanticipated as their electronic structure calls for electrons to pair up to form covalent bonds which results to a zero net magnetic moment, in contrast to natural ferromagnetic elements with unpaired electrons. Nevertheless, recent reports of magnetic ordering observed at room temperature for a number of different carbon allotropes such as highly oriented pyrolytic graphite (HOPG) [3] as well as graphene [4] have motivated new research studies in order to understand the intriguing mechanism(s) which can induce ferromagnetism in pure carbon systems. ⁎ Corresponding author at: School of Physics, University of the Witwatersrand, Johannesburg 2050, South Africa. E-mail address: [email protected] (N. Daya).
http://dx.doi.org/10.1016/j.diamond.2016.02.019 0925-9635/© 2016 Elsevier B.V. All rights reserved.
Research conducted by P. Esquinazi et al., has revealed ferromagnetism in proton irradiated highly oriented pyrolytic graphite (HOPG) with a Curie temperature above room temperature [3]. X-ray magnetic circular dichroism (XMCD) studies [5] performed by the same group on proton irradiated thin carbon films showed indisputably the ferromagnetic character of the irradiated regions and its relation to the carbon element. Since then, a number of experimental studies of magnetic carbon systems have confirmed similar observations. Consequently, a number of theoretical studies have been undertaken in order to explain the origin of carbon magnetism with respect to the presence and type of different defects in the graphene plane and in the graphite structure such as vacancies, edges and cracks which could nucleate magnetic moments and produce the observable effects with respect to magnetization and the associated Curie temperature [6,7]. We extend this research effort to the investigation of possible defect induced magnetic ordering in diamond (another allotrope of the carbon family) after proton irradiation. Diamond is a unique carbon allotrope which possesses exceptional and extreme properties such as hardness, electrical insulation, thermal conductivity, high charge carrier mobility and radiation hardness. As such diamond could find applications in microelectronics, spintronics, quantum cryptography, ultraviolet light-emitting diodes and optics, and high-power microwave electronics [8]. Ideal conditions for coherent spin manipulation can be found in diamond due to its long diffusion lengths and coherent times. We investigate proton irradiated nitrogen rich type Ib diamonds. The presence of nitrogen vacancy (NV) centres in diamond has sparked
198
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
much interest. The negatively charged NV centre has been shown to have weak spin–orbit coupling in the diamond matrix resulting in long spin coherence times at room temperature and long diffusion lengths. This allows for fast resonant spin manipulation. Coherent manipulation of individual electron spins associated with the NV colour centre using optical and microwave frequencies have already been demonstrated [9,10]. Numerous possibilities of the uses of magnetic carbon have been envisioned within this spintronics context and several technical challenges need to be overcome in order for these applications to be realised. One of the primary challenges is to induce magnetic ordering in these materials as a prerequisite for them being used as spin valves. In this work, we investigate the possibility of inducing magnetic ordering in nitrogen rich diamond type Ib through proton irradiation. This is the second generation of experiments performed for this purpose in diamond within our research group. In the first experiments ultrapure CVD grown type IIa diamonds were used [11]. The absence of π-electrons in the sp3 configuration of diamond could act as a test for theoretical calculations which have suggested that vacancy induced ferromagnetism is based on interaction between local magnetic moments and conduction π-electrons. In the present work, the emphasis lies on the theoretical suggestion that nitrogen resident nearby a vacancy can generate a larger magnetic signal as compared to a standalone carbon vacancy [7].
2. Experimental details Two nitrogen rich type Ib diamond samples were investigated for their magnetic properties. These diamonds were synthesised using high pressure, high temperature (HPHT) methods by the DeBeers Research laboratory (DRL) in South Africa and have a nitrogen content of approximately ~ 200 ppm. A mixed metal seed is used as a catalyst during the synthesis process which leads to magnetic impurities being contained within the samples. These magnetic impurities were quantified using Particle Induced X-Ray Emission (PIXE) at iThemba Labs Gauteng in South Africa and revealed a total magnetic impurity content of approximately 92.4 ppm. The two samples B1 and B2 have masses of 14.4 mg and 12.7 mg respectively, and dimensions ~3.3 × 1.9 × 0.66 mm3. Large area irradiations of the samples were conducted with 2.2 MeV protons using the 5 MV tandem accelerator at the Centre of Micro-Analysis of Materials (CMAM) at the Universidad Autónoma de Madrid. The proton beam was at normal incidence with respect to the diamond samples and parallel to the b001N axis. Sample B1 was irradiated with a total dose of approximately 1120 μC and a total fluence of 1.9 × 1017 H+/cm2. Sample B2 was irradiated with a total dose of approximately 4480 μC and a total fluence of 6.3 × 1017 H+/cm2 (See Table 1). Monte Carlo Simulations performed using the Stopping and Range of Ions in Matter (SRIM) [12] showed that protons with 2.2 MeV kinetic energy have a longitudinal range of approximately 28.5 μm in diamond and a longitudinal straggle of approximately 0.55 μm, with the displacement energy set to 45 eV [13] in order to account for the radiation hardness of diamond. Nuclear stopping is dominant near the range (~28.5 μm) of the protons and this is where the maximum defect density of 11 vacancies/ion on average are created according to SRIM calculations.
In order to investigate a change in the magnetic properties of the diamond samples, in terms of the proton induced defects and their associated magnetic moments, magnetisation measurements were conducted using a super conducting quantum interference device (SQUID) at the ICMM-CSIC in Madrid, Spain. These measurements were conducted on the pristine (before irradiation) samples as well as after the irradiation. Special care was taken in order to avoid magnetic contamination and in that respect the samples were mounted on specifically made gold-coated quartz sample holders using a small amount of diluted cryogenic varnish. The sample holders fit into the irradiation target chamber as well as the SQUID magnetometer [14]. This minimised the sample handling and allowed for the direct effects of the irradiation to be obtained, thus ensuring a high reproducibility of the observed results. The samples were further characterised using Raman Spectroscopy at the University of the Witwatersrand in South Africa. A 514.5 nm green emission line from an argon laser was utilised in conjunction with a Horiba Jobin-Yvon LabRAM HR Raman spectrometer. The laser beam was focused onto the samples using a microscope attachment with a 100× objective lens. The backscattered light was dispersed via a 600 line/mm grating onto a liquid nitrogen-cooled charge coupled device (CCD) detector. 3. Results and discussion Magnetisation measurements of the pristine samples at 300 K show that samples B1 and B2 exhibit a diamagnetic susceptibility of χ ≈ − 6.98 × 10−9 emu/Oe and χ ≈ − 6.42 × 10− 9 emu/Oe respectively. Hence, a diamagnetic susceptibility of the diamond samples of χ ≈ −4.85 × 10−7 emu/g and χ ≈ −5.05 × 10−7 emu/g is obtained respectively which correlates well with previous studies [15, 16]. SQUID measurements of the magnetic moment (μemu) of samples B1 and B2 as a function of the applied magnetic field (kOe) cycled between ± 50 kOe at 300 K are shown in Fig. 1. The linear diamagnetic background measured before the irradiations for the two samples has been subtracted respectively and an overall superparamagnetic behaviour can be conjectured. This superparamagnetic behaviour could be due to magnetic inclusions [17]. It should be recalled here that the growth of the studied crystals requires the use of iron as a catalyst and therefore, this can be the origin of the magnetic inclusions already detected in the pristine samples However, a closer look reveals a finite value for the coercive field (around 40 Oe) and a small non-zero remanence (around 2 × 10−7 emu) confirming an additional contribution of some blocked moments. It can therefore be concluded that there is a broad distribution of sizes in the magnetic inclusions; most of them are in a few nanometer length-scale and, consequently, they are
Table 1 Summary of the irradiation details for samples B1 and B2.
E (MeV) Qtotal (μC) Ftotal (H+/cm2) Beam area (mm2)
Sample B1
Sample B2
2.2 1120 1.9 × 1017 3.7
2.2 4480 6.3 × 1017 4.5
Fig. 1. Magnetic moments as a function of applied magnetic field at 300 K showing the superparamagnetic behaviour of samples B1 and B2.
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
199
unblocked and exhibit superparamagnetic behaviour. Only a small fraction of the magnetic moments, those corresponding to the larger size tail of the distribution, are blocked and contribute to the small ferromagnetic (FM) signal detected. Experiments exploring the ZFC (zero-field cooled) and FC (field-cooled) temperature dependence of the magnetization measured at low fields could help to assess a canonical paramagnetic behaviour and the corresponding blocking temperature but the extremely low signal near remanence (approximately 10−7 emu) superimposed onto a large diamagnetic background when measured at low fields, makes this strategy unfeasible. As it is apparent in Fig. 1, the saturation magnetization of sample B1 at 300 K is ~7.6 μemu and ~5.2 μemu for sample B2. The difference in the saturation is due to the slight difference in sample masses (with B2 being 1.7 mg lighter) resulting in a lower magnetic impurity content in B2. The saturation magnetisation of B2 remains essentially unchanged after irradiation yet a slight increase of ~ 2.3 μemu in the saturation magnetisation of B1 is observed. The magnetic effects of the irradiation are thus negligible at room temperature. Bharuth-Ram et al. have reported Mӧssbauer observations of hyperfine interaction signals in synthetic diamond which they speculated to be the superparamagnetic behaviour of nanoclustered impurities of Fe, which are introduced as a catalyst during the synthesis process [18]. To our knowledge, the present investigation confirms the Mossbauer results and it is the first observation of a behaviour that could be linked to the existence of superparamagnetism in synthetic diamond based on sensitive direct magnetometry of the magnetic moments as a function of applied field. Low temperature magnetisation measurements conducted at 4.2 K where thermal fluctuation effects are minimised are shown in Fig. 2. The low temperature measurements, along with the temperature dependence of the magnetization in Fig. 3, clearly indicate an increase
Fig. 3. Magnetic moment of diamond samples B1 and B2 as a function of temperature at 2 kOe on the negative constant diamagnetic background.
Fig. 2. Magnetic moments as a function of applied magnetic field at 4.2 K showing the increase in paramagnetic behaviour of samples B1 and B2.
in the paramagnetic behaviour of the samples after irradiation. This is directly related to the defects (vacancies) created during the irradiation. The saturation magnetisation of B1 is higher than that of B2 since B1 contains more ferromagnetic impurities. The overall system of defects created by irradiation remains paramagnetic with no evidence of a robust long range magnetic ordering at 4.2 K. Since the irradiated diamonds contain hydrogen in molecular and radical form, this result suggests that hydrogen might not be playing a role in giving rise to magnetic ordering in carbon, in agreement with several reports in literature. A careful examination of the thermal cycles shows that they exhibit hysteretic behaviour which is shown in Fig. 4. These measurements were repeated in order to verify their reproducibility and this hysteretic behaviour was always observed. Furthermore a transition temperature around 50–55 K was identified with the hysteretic behaviour occurring below this temperature in Figs. 3 and 4. The hysteretic behaviour observed and a weak tendency to saturate at high fields could also point to the existence of FM correlations in a highly disordered system. Thermal cycles were conducted at 2 kOe for temperatures between 4 K and 300 K which show that the pristine samples exhibit a Curietype paramagnetic behaviour. After subtraction of the magnetisation from the pristine samples, the net magnetic moment plotted versus the inverse temperature) exhibits linear magnetic behaviour during the cooling part of the thermal cycle confirming the formation of Curie-type paramagnetism in accordance with the Curie Law. By using a linear fit to the difference between the magnetisation of the irradiated and pristine measurements, a net value of the Curie constant of 0.0027 emu·K/g and 0.0037 emu·K/g is obtained for B1 and B2 respectively. By comparison of the Curie constants obtained from the linear fits
200
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
Fig. 5. Raman spectra of samples B1 and B2 before and after irradiation.
Fig. 4. Magnified region of the observed hysteretic behaviour in samples B1 and B2 after irradiation. Blue arrows indicate the cooling curves and red arrows indicate the heating curves.
to the pristine sample data, an effective paramagnetic increase of 22% for B1 and 30% for B2 was observed after the irradiation. The difference in the percentage increase in paramagnetic behaviour indicates a fluence dependence and hence a dependence on the amount of vacancies created during the irradiation. This observed behaviour allows for two possible explanations and both will be discussed. The behaviour can firstly be attributed to the single domain contributions from ferromagnetic impurities. However, if this postulate is to be believed, one would expect to see the same behaviour in the pristine samples. The pristine samples do not show the hysteresis observed with the irradiated samples thus suggesting that the observed behaviour is a result of the irradiation. The second explanation is that the observed behaviour is due to coupling of the irradiation induced amorphous carbon with the magnetic inclusions. Jin et al. reported an upturn of the magnetisation measurements in partially graphitized glassy carbon (GC) around 50 K [19]. They attributed this to amorphous carbon coupling with ferromagnetic clusters within the GC. Our hysteresis cycles very much resemble those of their graphitized glassy carbon. In their case, starting from a complete disordered system GC, they applied high temperature and pressures and the effect of order–disorder developed in the high pressures and high temperatures polymerisation result in topological defects that cause the occurrence of unpaired spins in this covalently bonded material. We are on the opposite side, starting from an ordered diamond, high levels of disorder are induced during the irradiation. The damage induced by irradiation has been assessed by Raman Spectroscopy as seen in Fig. 5. Raman spectra indicate that both pristine B1 and B2
samples display a clean diamond peak at 1333 cm−1. No other peaks are visible within the 0–2000 cm−1 range. After irradiation both the diamond peaks have a broader full width half maximum (FWHM) and B1 and B2 diamond peaks were downshifted by approximately 1 cm− 1. These changes are a result of the damaged induced during the irradiation [20]. The peaks within 1630–1640 cm− 1 range are attributed to the carbon interstitial related defect. The peak centred around the 1499 cm−1 for B2 is due to the vibrational modes of single vacancies. The peaks at 1365 and 1756 cm−1 roughly agree with the 3H optical centre [21]. In both cases, the order–disorder process would lead to a sp2–sp3 rehybridization and the forming of unpaired electrons that in turn could interact with the magnetic moments corresponding to the magnetic inclusions. Jin et al. hypothesised that the pressure induced graphitization could trigger an ordering alignment of the itinerant moments coming from these unpaired spins to render itinerant ferromagnetism. The upturn observed at 50 K in their data of M vs T was attributed to the coupling of paramagnetism of common amorphous carbon with some ferromagnetic clusters formed by defects. The dangling bonds in the amorphous state would be responsible for paramagnetism, and the itinerant spins may form in the polygonised (ordered inclusion) in glassy carbon. In our case, a larger hysteresis loop is observed in B1 which contains more ferromagnetic impurities. Therefore, an interaction of the moments in the implantation induced disordered areas with the magnetic moments in the magnetic impurities seems more probable. Furthermore, a similar study investigating proton irradiated ultra-pure type IIa diamond which have total magnetic impurity content of b 2 ppm and which were irradiated with a similar dose do not show the observed behaviour between 50 and 55 K [11], thus providing further support for the coupling discussed above.
4. Conclusion Magnetic impurities have caused much scepticism with regard to ferromagnetism observed in pure carbon materials. These results indicate that no ferromagnetism was observed in proton-irradiated type Ib diamonds despite the presence of magnetic impurities. Long range ordering of the magnetic inclusions did however produce a superparamagnetic signal which was observed at room temperature. The observed hysteretic behaviour below 55 K was attributed to the coupling of the amorphous carbon (created during the irradiation) and the ferromagnetic impurities contained within the sample.
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
Prime novelty statement No ferromagnetism was observed in proton-irradiated type Ib diamonds despite the presence of magnetic impurities. Long range ordering of the magnetic inclusions did however produce a superparamagnetic signal which was observed at room temperature. Acknowledgements This research was funded by the DST-NRF Centre of Excellence in Strong Materials, the National Research Foundation and the University of the Witwatersrand in South Africa. References [1] M. Coey, S. Sanvito, The magnetism of carbon, Phys. World 17 No 11 33–37 Polish Version Magetyzm Wegla, Postep. Fiz. 56 (2005) 122–127. [2] S. Talapatra, P.G. Ganesan, T. Kim, R. Vajtai, M. Huang, M. Shima, et al., Irradiationinduced magnetism in carbon nanostructures, Phys. Rev. Lett. 95 (2005) 15–18, http://dx.doi.org/10.1103/PhysRevLett.95.097201. [3] P. Esquinazi, D. Spemann, R. Höhne, A. Setzer, K.-H. Han, T. Butz, Induced magnetic ordering by proton irradiation in graphite, Phys. Rev. Lett. 91 (2003) 227201, http:// dx.doi.org/10.1103/PhysRevLett.91.227201. [4] Y. Wang, Y. Hoang, Y. Song, X. Zhang, Y. Ma, J. Liang, et al., Room-temperature ferromagnetism of graphene, Nano Lett. 9 (2009) 220–224, http://dx.doi.org/10.1021/ nl802810g. [5] H. Ohldag, T. Tyliszczak, R. Höhne, D. Spemann, P. Esquinazi, M. Ungureanu, et al., Πelectron ferromagnetism in metal-free carbon probed by soft X-ray dichroism, Phys. Rev. Lett. 98 (2007) 1–4, http://dx.doi.org/10.1103/PhysRevLett.98.187204. [6] X. Yang, H. Xia, X. Qin, W. Li, Y. Dai, X. Liu, et al., Correlation between the vacancy defects and ferromagnetism in graphite, Carbon N. Y. 47 (2009) 1399–1406, http://dx.doi.org/10.1016/j.carbon.2009.01.032. [7] Y. Zhang, S. Talapatra, S. Kar, R. Vajtai, S.K. Nayak, P.M. Ajayan, First-principles study of defect-induced magnetism in carbon, Phys. Rev. Lett. 99 (2007) 107201, http:// dx.doi.org/10.1103/PhysRevLett.99.107201. [8] B.D.D. Awschalom, R. Hanson, Diamond age, Sci. Am. (2007).
201
[9] M.V.G. Dutt, L. Childress, L. Jiang, E. Togan, J. Maze, F. Jelezko, et al., Quantum register based on individual electronic and nuclear spin qubits in diamond, Science (80-.) 316 (2007) 1312–1316, http://dx.doi.org/10.1126/science.1139831. [10] R. Hanson, O. Gywat, D.D. Awschalom, Room-temperature manipulation and decoherence of a single spin in diamond, Phys. Rev. B Condens. Matter Mater. Phys. 74 (2006) 1–4, http://dx.doi.org/10.1103/PhysRevB.74.161203. [11] T. Makgato, To be published, n.d. [12] J.F. Ziegler, M.D. Ziegler, J.P. Biersack, SRIM — the stopping and range of ions in matter (2010), Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 268 (2010) 1818–1823, http://dx.doi.org/10.1016/j.nimb.2010.02.091. [13] E. Friedland, H.D. Carstanjen, G. Myburg, M.A. Nasr, Dependence of damage efficiency of ions in diamond on electronic stopping, Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 230 (2005) 129–135, http://dx.doi.org/10.1016/j. nimb.2004.12.030. [14] M.A. Ramos, J. Barzola-Quiquia, P. Esquinazi, A. Muñoz-Martin, A. Climent-Font, M. García-Hernández, Magnetic properties of graphite irradiated with MeV ions, Phys. Rev. B 81 (2010) 214404, http://dx.doi.org/10.1103/PhysRevB.81.214404. [15] T. Makarova, Magnetic properties of carbon structures, Semiconductors 38 (2004) 615–638, http://dx.doi.org/10.1134/1.1766362. [16] J. Heremans, C.H. Olk, D.T. Morelli, Magnetic susceptibility of carbon structures, Phys. Rev. B 49 (1994) 15122–15125, http://dx.doi.org/10.1103/PhysRevB.49. 15122. [17] H.G. Maguire, Electron and nuclear structural characterization of natural, synthetic, homoepitaxial and polycrystalline low pressure chemically vapour- deposited diamond, Mater. Sci. (1992) 243–248. [18] K. Bharuth-Ram, H. Haricharun, N. Govender, Mössbauer study of Fe inclusions in synthetic diamond, Hyperfine Interact. 71 (1992) 1407–1410 (http://www. springerlink.com/index/9q6q16g471732704.pdf). [19] C. Jin, X. Wang, Z. Liu, Y. Zhang, The unusual morphology, structure, and magnetic property evolution of glassy carbon upon high pressure treatment, Braz. J. 33 (2003) 723–728 (http://www.scielo.br/scielo.php?pid=S0103-97332003000400017&script=sci_ arttext&tlng=es). [20] J. Orwa, K. Nugent, D. Jamieson, S. Prawer, Raman investigation of damage caused by deep ion implantation in diamond, Phys. Rev. B 62 (2000) 5461–5472, http://dx.doi. org/10.1103/PhysRevB.62.5461. [21] S. Prawer, I. Rosenblum, J.O. Orwa, J. Adler, Identification of the point defects in diamond as measured by Raman spectroscopy: comparison between experiment and computation, Chem. Phys. Lett. 390 (2004) 458–461, http://dx.doi.org/10. 1016/j.cplett.2004.04.027.
Contents lists available at ScienceDirect
Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond
Investigation of the magnetic properties of proton irradiated type Ib HPHT diamond N. Daya a,e,⁎, E. Sideras-Haddad a,e, T.N. Makgato a,e, M. García-Hernández b, A. Climent-Font c,d, A. Zucchiatti c, M.A. Ramos c,d a
School of Physics, University of the Witwatersrand, Johannesburg 2050, South Africa Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, E-28049 Madrid, Spain Centro de Microanálisis de Materiales (CMAM), Universidad Autónoma de Madrid, Cantoblanco, E-28049 Madrid, Spain d Instituto de Ciencia de Materiales “Nicolás Cabrera”, Universidad Autónoma de Madrid, Cantoblanco, E-28049 Madrid, Spain e Centre of Excellence in Strong Materials, Physics Building, University of the Witwatersrand, Johannesburg 2050, South Africa b c
a r t i c l e
i n f o
Article history: Received 1 November 2015 Received in revised form 22 February 2016 Accepted 23 February 2016 Available online 26 February 2016 Keywords: Amorphous carbon Diamond Magnetism
a b s t r a c t Nitrogen rich type Ib HPHT grown synthetic diamonds were investigated following large area irradiation of the samples using 2.2 MeV protons. Studies on possible magnetic properties induced after the irradiation were performed using a SQUID magnetometer. Magnetisation measurements of pristine (unirradiated) control samples revealed a superparamagnetic-like signal at 300 K. After the proton irradiation, a Curie-like paramagnetic curve was observed for thermal cycles at an applied magnetic field of 2 kOe which exhibits a transition at temperatures around 50–55 K, with hysteretic behaviour below these temperatures. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The magnetic properties of carbon have gained significant attention over the recent years with extensive experimental and theoretical research continuously been performed on this challenging topic. Magnetic ordering in this class of carbon-based materials at room temperatures could lead to revolutionary new technologies in nanotechnology, biomedicine and many other fields [1] since they are light weight and inexpensive to produce [2]. Such materials present unique prospects in terms of spin-based electronics due to the fact that the weak spin–orbit coupling in carbon results in the critically desired long diffusion lengths and coherence times. However, ferromagnetism in carbon allotropes is rather unanticipated as their electronic structure calls for electrons to pair up to form covalent bonds which results to a zero net magnetic moment, in contrast to natural ferromagnetic elements with unpaired electrons. Nevertheless, recent reports of magnetic ordering observed at room temperature for a number of different carbon allotropes such as highly oriented pyrolytic graphite (HOPG) [3] as well as graphene [4] have motivated new research studies in order to understand the intriguing mechanism(s) which can induce ferromagnetism in pure carbon systems. ⁎ Corresponding author at: School of Physics, University of the Witwatersrand, Johannesburg 2050, South Africa. E-mail address: [email protected] (N. Daya).
http://dx.doi.org/10.1016/j.diamond.2016.02.019 0925-9635/© 2016 Elsevier B.V. All rights reserved.
Research conducted by P. Esquinazi et al., has revealed ferromagnetism in proton irradiated highly oriented pyrolytic graphite (HOPG) with a Curie temperature above room temperature [3]. X-ray magnetic circular dichroism (XMCD) studies [5] performed by the same group on proton irradiated thin carbon films showed indisputably the ferromagnetic character of the irradiated regions and its relation to the carbon element. Since then, a number of experimental studies of magnetic carbon systems have confirmed similar observations. Consequently, a number of theoretical studies have been undertaken in order to explain the origin of carbon magnetism with respect to the presence and type of different defects in the graphene plane and in the graphite structure such as vacancies, edges and cracks which could nucleate magnetic moments and produce the observable effects with respect to magnetization and the associated Curie temperature [6,7]. We extend this research effort to the investigation of possible defect induced magnetic ordering in diamond (another allotrope of the carbon family) after proton irradiation. Diamond is a unique carbon allotrope which possesses exceptional and extreme properties such as hardness, electrical insulation, thermal conductivity, high charge carrier mobility and radiation hardness. As such diamond could find applications in microelectronics, spintronics, quantum cryptography, ultraviolet light-emitting diodes and optics, and high-power microwave electronics [8]. Ideal conditions for coherent spin manipulation can be found in diamond due to its long diffusion lengths and coherent times. We investigate proton irradiated nitrogen rich type Ib diamonds. The presence of nitrogen vacancy (NV) centres in diamond has sparked
198
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
much interest. The negatively charged NV centre has been shown to have weak spin–orbit coupling in the diamond matrix resulting in long spin coherence times at room temperature and long diffusion lengths. This allows for fast resonant spin manipulation. Coherent manipulation of individual electron spins associated with the NV colour centre using optical and microwave frequencies have already been demonstrated [9,10]. Numerous possibilities of the uses of magnetic carbon have been envisioned within this spintronics context and several technical challenges need to be overcome in order for these applications to be realised. One of the primary challenges is to induce magnetic ordering in these materials as a prerequisite for them being used as spin valves. In this work, we investigate the possibility of inducing magnetic ordering in nitrogen rich diamond type Ib through proton irradiation. This is the second generation of experiments performed for this purpose in diamond within our research group. In the first experiments ultrapure CVD grown type IIa diamonds were used [11]. The absence of π-electrons in the sp3 configuration of diamond could act as a test for theoretical calculations which have suggested that vacancy induced ferromagnetism is based on interaction between local magnetic moments and conduction π-electrons. In the present work, the emphasis lies on the theoretical suggestion that nitrogen resident nearby a vacancy can generate a larger magnetic signal as compared to a standalone carbon vacancy [7].
2. Experimental details Two nitrogen rich type Ib diamond samples were investigated for their magnetic properties. These diamonds were synthesised using high pressure, high temperature (HPHT) methods by the DeBeers Research laboratory (DRL) in South Africa and have a nitrogen content of approximately ~ 200 ppm. A mixed metal seed is used as a catalyst during the synthesis process which leads to magnetic impurities being contained within the samples. These magnetic impurities were quantified using Particle Induced X-Ray Emission (PIXE) at iThemba Labs Gauteng in South Africa and revealed a total magnetic impurity content of approximately 92.4 ppm. The two samples B1 and B2 have masses of 14.4 mg and 12.7 mg respectively, and dimensions ~3.3 × 1.9 × 0.66 mm3. Large area irradiations of the samples were conducted with 2.2 MeV protons using the 5 MV tandem accelerator at the Centre of Micro-Analysis of Materials (CMAM) at the Universidad Autónoma de Madrid. The proton beam was at normal incidence with respect to the diamond samples and parallel to the b001N axis. Sample B1 was irradiated with a total dose of approximately 1120 μC and a total fluence of 1.9 × 1017 H+/cm2. Sample B2 was irradiated with a total dose of approximately 4480 μC and a total fluence of 6.3 × 1017 H+/cm2 (See Table 1). Monte Carlo Simulations performed using the Stopping and Range of Ions in Matter (SRIM) [12] showed that protons with 2.2 MeV kinetic energy have a longitudinal range of approximately 28.5 μm in diamond and a longitudinal straggle of approximately 0.55 μm, with the displacement energy set to 45 eV [13] in order to account for the radiation hardness of diamond. Nuclear stopping is dominant near the range (~28.5 μm) of the protons and this is where the maximum defect density of 11 vacancies/ion on average are created according to SRIM calculations.
In order to investigate a change in the magnetic properties of the diamond samples, in terms of the proton induced defects and their associated magnetic moments, magnetisation measurements were conducted using a super conducting quantum interference device (SQUID) at the ICMM-CSIC in Madrid, Spain. These measurements were conducted on the pristine (before irradiation) samples as well as after the irradiation. Special care was taken in order to avoid magnetic contamination and in that respect the samples were mounted on specifically made gold-coated quartz sample holders using a small amount of diluted cryogenic varnish. The sample holders fit into the irradiation target chamber as well as the SQUID magnetometer [14]. This minimised the sample handling and allowed for the direct effects of the irradiation to be obtained, thus ensuring a high reproducibility of the observed results. The samples were further characterised using Raman Spectroscopy at the University of the Witwatersrand in South Africa. A 514.5 nm green emission line from an argon laser was utilised in conjunction with a Horiba Jobin-Yvon LabRAM HR Raman spectrometer. The laser beam was focused onto the samples using a microscope attachment with a 100× objective lens. The backscattered light was dispersed via a 600 line/mm grating onto a liquid nitrogen-cooled charge coupled device (CCD) detector. 3. Results and discussion Magnetisation measurements of the pristine samples at 300 K show that samples B1 and B2 exhibit a diamagnetic susceptibility of χ ≈ − 6.98 × 10−9 emu/Oe and χ ≈ − 6.42 × 10− 9 emu/Oe respectively. Hence, a diamagnetic susceptibility of the diamond samples of χ ≈ −4.85 × 10−7 emu/g and χ ≈ −5.05 × 10−7 emu/g is obtained respectively which correlates well with previous studies [15, 16]. SQUID measurements of the magnetic moment (μemu) of samples B1 and B2 as a function of the applied magnetic field (kOe) cycled between ± 50 kOe at 300 K are shown in Fig. 1. The linear diamagnetic background measured before the irradiations for the two samples has been subtracted respectively and an overall superparamagnetic behaviour can be conjectured. This superparamagnetic behaviour could be due to magnetic inclusions [17]. It should be recalled here that the growth of the studied crystals requires the use of iron as a catalyst and therefore, this can be the origin of the magnetic inclusions already detected in the pristine samples However, a closer look reveals a finite value for the coercive field (around 40 Oe) and a small non-zero remanence (around 2 × 10−7 emu) confirming an additional contribution of some blocked moments. It can therefore be concluded that there is a broad distribution of sizes in the magnetic inclusions; most of them are in a few nanometer length-scale and, consequently, they are
Table 1 Summary of the irradiation details for samples B1 and B2.
E (MeV) Qtotal (μC) Ftotal (H+/cm2) Beam area (mm2)
Sample B1
Sample B2
2.2 1120 1.9 × 1017 3.7
2.2 4480 6.3 × 1017 4.5
Fig. 1. Magnetic moments as a function of applied magnetic field at 300 K showing the superparamagnetic behaviour of samples B1 and B2.
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
199
unblocked and exhibit superparamagnetic behaviour. Only a small fraction of the magnetic moments, those corresponding to the larger size tail of the distribution, are blocked and contribute to the small ferromagnetic (FM) signal detected. Experiments exploring the ZFC (zero-field cooled) and FC (field-cooled) temperature dependence of the magnetization measured at low fields could help to assess a canonical paramagnetic behaviour and the corresponding blocking temperature but the extremely low signal near remanence (approximately 10−7 emu) superimposed onto a large diamagnetic background when measured at low fields, makes this strategy unfeasible. As it is apparent in Fig. 1, the saturation magnetization of sample B1 at 300 K is ~7.6 μemu and ~5.2 μemu for sample B2. The difference in the saturation is due to the slight difference in sample masses (with B2 being 1.7 mg lighter) resulting in a lower magnetic impurity content in B2. The saturation magnetisation of B2 remains essentially unchanged after irradiation yet a slight increase of ~ 2.3 μemu in the saturation magnetisation of B1 is observed. The magnetic effects of the irradiation are thus negligible at room temperature. Bharuth-Ram et al. have reported Mӧssbauer observations of hyperfine interaction signals in synthetic diamond which they speculated to be the superparamagnetic behaviour of nanoclustered impurities of Fe, which are introduced as a catalyst during the synthesis process [18]. To our knowledge, the present investigation confirms the Mossbauer results and it is the first observation of a behaviour that could be linked to the existence of superparamagnetism in synthetic diamond based on sensitive direct magnetometry of the magnetic moments as a function of applied field. Low temperature magnetisation measurements conducted at 4.2 K where thermal fluctuation effects are minimised are shown in Fig. 2. The low temperature measurements, along with the temperature dependence of the magnetization in Fig. 3, clearly indicate an increase
Fig. 3. Magnetic moment of diamond samples B1 and B2 as a function of temperature at 2 kOe on the negative constant diamagnetic background.
Fig. 2. Magnetic moments as a function of applied magnetic field at 4.2 K showing the increase in paramagnetic behaviour of samples B1 and B2.
in the paramagnetic behaviour of the samples after irradiation. This is directly related to the defects (vacancies) created during the irradiation. The saturation magnetisation of B1 is higher than that of B2 since B1 contains more ferromagnetic impurities. The overall system of defects created by irradiation remains paramagnetic with no evidence of a robust long range magnetic ordering at 4.2 K. Since the irradiated diamonds contain hydrogen in molecular and radical form, this result suggests that hydrogen might not be playing a role in giving rise to magnetic ordering in carbon, in agreement with several reports in literature. A careful examination of the thermal cycles shows that they exhibit hysteretic behaviour which is shown in Fig. 4. These measurements were repeated in order to verify their reproducibility and this hysteretic behaviour was always observed. Furthermore a transition temperature around 50–55 K was identified with the hysteretic behaviour occurring below this temperature in Figs. 3 and 4. The hysteretic behaviour observed and a weak tendency to saturate at high fields could also point to the existence of FM correlations in a highly disordered system. Thermal cycles were conducted at 2 kOe for temperatures between 4 K and 300 K which show that the pristine samples exhibit a Curietype paramagnetic behaviour. After subtraction of the magnetisation from the pristine samples, the net magnetic moment plotted versus the inverse temperature) exhibits linear magnetic behaviour during the cooling part of the thermal cycle confirming the formation of Curie-type paramagnetism in accordance with the Curie Law. By using a linear fit to the difference between the magnetisation of the irradiated and pristine measurements, a net value of the Curie constant of 0.0027 emu·K/g and 0.0037 emu·K/g is obtained for B1 and B2 respectively. By comparison of the Curie constants obtained from the linear fits
200
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
Fig. 5. Raman spectra of samples B1 and B2 before and after irradiation.
Fig. 4. Magnified region of the observed hysteretic behaviour in samples B1 and B2 after irradiation. Blue arrows indicate the cooling curves and red arrows indicate the heating curves.
to the pristine sample data, an effective paramagnetic increase of 22% for B1 and 30% for B2 was observed after the irradiation. The difference in the percentage increase in paramagnetic behaviour indicates a fluence dependence and hence a dependence on the amount of vacancies created during the irradiation. This observed behaviour allows for two possible explanations and both will be discussed. The behaviour can firstly be attributed to the single domain contributions from ferromagnetic impurities. However, if this postulate is to be believed, one would expect to see the same behaviour in the pristine samples. The pristine samples do not show the hysteresis observed with the irradiated samples thus suggesting that the observed behaviour is a result of the irradiation. The second explanation is that the observed behaviour is due to coupling of the irradiation induced amorphous carbon with the magnetic inclusions. Jin et al. reported an upturn of the magnetisation measurements in partially graphitized glassy carbon (GC) around 50 K [19]. They attributed this to amorphous carbon coupling with ferromagnetic clusters within the GC. Our hysteresis cycles very much resemble those of their graphitized glassy carbon. In their case, starting from a complete disordered system GC, they applied high temperature and pressures and the effect of order–disorder developed in the high pressures and high temperatures polymerisation result in topological defects that cause the occurrence of unpaired spins in this covalently bonded material. We are on the opposite side, starting from an ordered diamond, high levels of disorder are induced during the irradiation. The damage induced by irradiation has been assessed by Raman Spectroscopy as seen in Fig. 5. Raman spectra indicate that both pristine B1 and B2
samples display a clean diamond peak at 1333 cm−1. No other peaks are visible within the 0–2000 cm−1 range. After irradiation both the diamond peaks have a broader full width half maximum (FWHM) and B1 and B2 diamond peaks were downshifted by approximately 1 cm− 1. These changes are a result of the damaged induced during the irradiation [20]. The peaks within 1630–1640 cm− 1 range are attributed to the carbon interstitial related defect. The peak centred around the 1499 cm−1 for B2 is due to the vibrational modes of single vacancies. The peaks at 1365 and 1756 cm−1 roughly agree with the 3H optical centre [21]. In both cases, the order–disorder process would lead to a sp2–sp3 rehybridization and the forming of unpaired electrons that in turn could interact with the magnetic moments corresponding to the magnetic inclusions. Jin et al. hypothesised that the pressure induced graphitization could trigger an ordering alignment of the itinerant moments coming from these unpaired spins to render itinerant ferromagnetism. The upturn observed at 50 K in their data of M vs T was attributed to the coupling of paramagnetism of common amorphous carbon with some ferromagnetic clusters formed by defects. The dangling bonds in the amorphous state would be responsible for paramagnetism, and the itinerant spins may form in the polygonised (ordered inclusion) in glassy carbon. In our case, a larger hysteresis loop is observed in B1 which contains more ferromagnetic impurities. Therefore, an interaction of the moments in the implantation induced disordered areas with the magnetic moments in the magnetic impurities seems more probable. Furthermore, a similar study investigating proton irradiated ultra-pure type IIa diamond which have total magnetic impurity content of b 2 ppm and which were irradiated with a similar dose do not show the observed behaviour between 50 and 55 K [11], thus providing further support for the coupling discussed above.
4. Conclusion Magnetic impurities have caused much scepticism with regard to ferromagnetism observed in pure carbon materials. These results indicate that no ferromagnetism was observed in proton-irradiated type Ib diamonds despite the presence of magnetic impurities. Long range ordering of the magnetic inclusions did however produce a superparamagnetic signal which was observed at room temperature. The observed hysteretic behaviour below 55 K was attributed to the coupling of the amorphous carbon (created during the irradiation) and the ferromagnetic impurities contained within the sample.
N. Daya et al. / Diamond & Related Materials 64 (2016) 197–201
Prime novelty statement No ferromagnetism was observed in proton-irradiated type Ib diamonds despite the presence of magnetic impurities. Long range ordering of the magnetic inclusions did however produce a superparamagnetic signal which was observed at room temperature. Acknowledgements This research was funded by the DST-NRF Centre of Excellence in Strong Materials, the National Research Foundation and the University of the Witwatersrand in South Africa. References [1] M. Coey, S. Sanvito, The magnetism of carbon, Phys. World 17 No 11 33–37 Polish Version Magetyzm Wegla, Postep. Fiz. 56 (2005) 122–127. [2] S. Talapatra, P.G. Ganesan, T. Kim, R. Vajtai, M. Huang, M. Shima, et al., Irradiationinduced magnetism in carbon nanostructures, Phys. Rev. Lett. 95 (2005) 15–18, http://dx.doi.org/10.1103/PhysRevLett.95.097201. [3] P. Esquinazi, D. Spemann, R. Höhne, A. Setzer, K.-H. Han, T. Butz, Induced magnetic ordering by proton irradiation in graphite, Phys. Rev. Lett. 91 (2003) 227201, http:// dx.doi.org/10.1103/PhysRevLett.91.227201. [4] Y. Wang, Y. Hoang, Y. Song, X. Zhang, Y. Ma, J. Liang, et al., Room-temperature ferromagnetism of graphene, Nano Lett. 9 (2009) 220–224, http://dx.doi.org/10.1021/ nl802810g. [5] H. Ohldag, T. Tyliszczak, R. Höhne, D. Spemann, P. Esquinazi, M. Ungureanu, et al., Πelectron ferromagnetism in metal-free carbon probed by soft X-ray dichroism, Phys. Rev. Lett. 98 (2007) 1–4, http://dx.doi.org/10.1103/PhysRevLett.98.187204. [6] X. Yang, H. Xia, X. Qin, W. Li, Y. Dai, X. Liu, et al., Correlation between the vacancy defects and ferromagnetism in graphite, Carbon N. Y. 47 (2009) 1399–1406, http://dx.doi.org/10.1016/j.carbon.2009.01.032. [7] Y. Zhang, S. Talapatra, S. Kar, R. Vajtai, S.K. Nayak, P.M. Ajayan, First-principles study of defect-induced magnetism in carbon, Phys. Rev. Lett. 99 (2007) 107201, http:// dx.doi.org/10.1103/PhysRevLett.99.107201. [8] B.D.D. Awschalom, R. Hanson, Diamond age, Sci. Am. (2007).
201
[9] M.V.G. Dutt, L. Childress, L. Jiang, E. Togan, J. Maze, F. Jelezko, et al., Quantum register based on individual electronic and nuclear spin qubits in diamond, Science (80-.) 316 (2007) 1312–1316, http://dx.doi.org/10.1126/science.1139831. [10] R. Hanson, O. Gywat, D.D. Awschalom, Room-temperature manipulation and decoherence of a single spin in diamond, Phys. Rev. B Condens. Matter Mater. Phys. 74 (2006) 1–4, http://dx.doi.org/10.1103/PhysRevB.74.161203. [11] T. Makgato, To be published, n.d. [12] J.F. Ziegler, M.D. Ziegler, J.P. Biersack, SRIM — the stopping and range of ions in matter (2010), Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 268 (2010) 1818–1823, http://dx.doi.org/10.1016/j.nimb.2010.02.091. [13] E. Friedland, H.D. Carstanjen, G. Myburg, M.A. Nasr, Dependence of damage efficiency of ions in diamond on electronic stopping, Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 230 (2005) 129–135, http://dx.doi.org/10.1016/j. nimb.2004.12.030. [14] M.A. Ramos, J. Barzola-Quiquia, P. Esquinazi, A. Muñoz-Martin, A. Climent-Font, M. García-Hernández, Magnetic properties of graphite irradiated with MeV ions, Phys. Rev. B 81 (2010) 214404, http://dx.doi.org/10.1103/PhysRevB.81.214404. [15] T. Makarova, Magnetic properties of carbon structures, Semiconductors 38 (2004) 615–638, http://dx.doi.org/10.1134/1.1766362. [16] J. Heremans, C.H. Olk, D.T. Morelli, Magnetic susceptibility of carbon structures, Phys. Rev. B 49 (1994) 15122–15125, http://dx.doi.org/10.1103/PhysRevB.49. 15122. [17] H.G. Maguire, Electron and nuclear structural characterization of natural, synthetic, homoepitaxial and polycrystalline low pressure chemically vapour- deposited diamond, Mater. Sci. (1992) 243–248. [18] K. Bharuth-Ram, H. Haricharun, N. Govender, Mössbauer study of Fe inclusions in synthetic diamond, Hyperfine Interact. 71 (1992) 1407–1410 (http://www. springerlink.com/index/9q6q16g471732704.pdf). [19] C. Jin, X. Wang, Z. Liu, Y. Zhang, The unusual morphology, structure, and magnetic property evolution of glassy carbon upon high pressure treatment, Braz. J. 33 (2003) 723–728 (http://www.scielo.br/scielo.php?pid=S0103-97332003000400017&script=sci_ arttext&tlng=es). [20] J. Orwa, K. Nugent, D. Jamieson, S. Prawer, Raman investigation of damage caused by deep ion implantation in diamond, Phys. Rev. B 62 (2000) 5461–5472, http://dx.doi. org/10.1103/PhysRevB.62.5461. [21] S. Prawer, I. Rosenblum, J.O. Orwa, J. Adler, Identification of the point defects in diamond as measured by Raman spectroscopy: comparison between experiment and computation, Chem. Phys. Lett. 390 (2004) 458–461, http://dx.doi.org/10. 1016/j.cplett.2004.04.027.