Full Length Article
The role of atomic oxygen in the decomposition of self-assembled monolayers during area-selective atomic layer deposition

https://doi.org/10.1016/j.apsusc.2022.152679Get rights and content

Highlights

  • Atomic oxygen exposure mimics oxidation pulse experienced during atomic layer deposition.

  • A short chain DETA SAM with amino terminal group reveals a linear etch rate with successive atomic oxygen exposures.

  • Etch rate is not dependent on the exposure size, small and large exposures etch the DETA at the same rate.

  • A long chain OTMS SAM with no terminal group reveals an exponential etch rate with successive atomic oxygen exposures.

Abstract

Utilising self-assembled monolayers (SAMs) to achieve area-selective atomic layer deposition (AS-ALD) as an approach to bottom-up nanofabrication has recently gained significant attention from the nanoelectronics industry. With the continued downscaling of feature sizes, top-down processing can no longer reach the challenging demands of the industry which requires conformal coating of high aspect ratio vias and a reduction in misalignment errors in multi-layered devices. In this work we attempt to imitate the effects of the ALD oxidation pulse experienced by the SAMs during the AS-ALD process by exposing two SAMs of different chain lengths and different functional groups, (3-trimethoxysilylpropyl)diethylenetriamine (DETA) and octadecyltrimethoxysilane (OTMS), to numerous controlled in-vacuo atomic oxygen exposures with subsequent characterisation by X-ray photoelectron spectroscopy (XPS). We monitor the sequential removal of the deposited monolayers with each successive atomic oxygen exposure for both SAMs. The etch rate is observed to be distinct for the different SAMs, the amino-terminated short chain DETA SAM reveals a linear etch rate while the longer chain OTMS SAM reveals an exponential etch rate. The results presented provide some insights into what characteristics are important for choosing the correct SAM for AS-ALD applications.

Introduction

Atomic Layer Deposition (ALD) has greatly enhanced the capability to deposit very uniform and conformal thin films on high aspect ratio structures allowing for the continued downscaling of feature sizes in nanoelectronic devices. ALD is largely compatible with semiconductor device fabrication processes and has many other advantages that make it an attractive deposition method [1], [2], [3]. One main drawback for ALD is that the same conformality that leads to uniform deposition across large substrates, leads to increased process complexity when patterning of the surface is required. Area selective ALD (AS-ALD) is proposed as a method for preferentially depositing a material on a specific area on the surface while the remainder is left uncoated, giving control over a deposited thin film in the horizontal plane[4], [5], [6], [7], [8]. One approach being adopted to achieve area selective ALD is through the use of self-assembled monolayers (SAMs) which can assist in the patterning of microscale and nanoscale features [9], [10], [11], [12], [13], [14], [15], [16]. The adaptability of the SAM chemistry makes them a prime candidate for selectively adhering to particular sections of the surface allowing them to block any subsequently deposited material. Herregods et al. describe a bottom-up manufacturing process for electronic devices in the back-end-of-line (BEOL) which consists of using SAMs to passivate the surface and consequently block film growth by ALD [17]. Device geometries in the BEOL typically consist of alternating patterns of metal and dielectric material on the wafer surface and the process methodology often calls for deposition only on either the metal or the dielectric. In the case of deposition on the metallic surface, a silane-based SAM can be deposited on the dielectric surface and by choosing an appropriate terminal functional group it will block the subsequent deposited material from bonding to the dielectric. In this way the material is only deposited on the metal and not the dielectric [18], [19], [20]. Conversely, using thiol [21] or phosphonic acid-based [22] SAMs results in material being deposited on the dielectric and not the metal/metal oxide. The SAM is then removed by etching [23], [24] or electrochemically by applying a potential bias [25]. The area selective process is schematically displayed in Fig. 1 (a).

This method can also be used for depositing on 3D nanostructures. Chopra et al. have shown how AS-ALD can differ on planar surfaces and non-planar surfaces. They were able to selectively block TiN deposition on hafnium dioxide (HfO2) on the planar surfaces but grow TiN films on the nano-lines and nano-pillars where the SAM film had difficulty creating a uniform layer [27]. Dong et al. have also demonstrated their AS-ALD technique for successfully coating the vertical surfaces of nanopillars with zinc oxide (ZnO) [28]. Mameli et al. have introduced a cyclic style AS-ALD process where inhibitor molecules are reapplied after each cycle [29]. The deposition of dielectrics by ALD, both on metals and dielectric substrates, typically requires oxygen as a co-reactant derived either from water, for example in the trimethylaluminum-based Al2O3 process [30], or an oxygen plasma during plasma-enhanced ALD (PE-ALD)[31]. Although the use of plasma allows for a greater range of materials to be deposited, at lower temperatures the films are not as conformal or uniform, and they are known to cause damage to the underlying layers [31], [32], [33]. Lee et al. have shown how the choice of plasma during AS-ALD can degrade a SAM film up to the point where no selective deposition is observed. By changing their gaseous plasma from an NH3 to a H2 they successfully deposited thin cobalt films up to 1000 cycles [34]. An O2 plasma is frequently used as the reactant plasma for the growth of a range of materials including, Al2O3, HfO2, SiO2, Pt, Ru, TiO2 and ZnO [31], [32], [35] and more recently, molecular oxygen has been utilized for metal oxide deposition [36]. However, O2 plasmas have been shown to cause degradation of low-k dielectric films due to ions and oxygen radicals causing damage to the dielectric [37]. This greatly affects the device performance and circuit reliability. With regards to SAMs, George et al. have shown how an O2 plasma can be used to etch a 1-octadecanethiol (ODT) SAM and remove it completely from the surface. In their study the SAM, which was deposited on a gold covered Si substrate, was partially covered by a poly(dimethylsiloxane) (PDMS) stamp. Subsequently, an O2 plasma was used to etch the SAM from regions that were not covered by the PDMS stamp, creating a pattern. SEM and AFM performed after the electrodeposition of ZnO, Ni and Ag onto the patterned samples, found that these materials were only deposited on the underlying gold, while the SAM effectively blocked the deposition of these materials [38].

Atomic oxygen has been used as an ion-free alternative to O2 plasma as it has the capacity to be less aggressive because it is a low-energy treatment that is also electrically neutral. Chaudhari et al. have assessed the role atomic oxygen plays in O2 plasma induced damage of ultralow-k dielectrics [39]. Bogan et al. have used atomic oxygen as a less aggressive replacement for plasma treatment to modify dense low-k dielectric surfaces [40]. Dai et al. using self-assembled monolayers as a substitute for wool fibres, investigated the impact of atomic oxygen and O2 plasma on an alkylthiolate SAM. They found that the plasma treatment had a more significant effect and etched the SAM much more rapidly compared to atomic oxygen treatment [41].

In an attempt to understand the contribution of atomic oxygen to the decomposition of self-assembled monolayers, the current study examines the impact of controlled atomic oxygen exposures on two SAMs which are of interest for AS-ALD applications. This atomic oxygen treatment essentially mimics the oxidation pulse performed during the ALD cycle without the presence of other species. The stepwise treatment of a (3-trimethoxysilylpropyl) diethylenetriamine (DETA) SAM (Fig. 1(b)) is highly controlled using successive 100 Langmuir (L) exposures and examined using XPS measurements. The impact of these small atomic oxygen exposures is compared with much larger 1000 L exposures, and the changes induced on the DETA SAM terminated SiO2 substrate is examined. The same process is followed for the carbon-terminated octadecyltrimethoxysilane (OTMS) SAM (Fig. 1(b)), although larger 500 L exposures are used. The etch rate resulting from these oxidation steps for each SAM is studied.

Section snippets

Experimental details

In this work two different SAMs were deposited on Si wafers with a native silicon dioxide layer and subsequently characterised by XPS. The samples were then treated with controlled atomic oxygen exposures, with XPS spectra recorded after each exposure. The Si substrates were initially cleaned by UV-ozone treatment for 15 mins in a Jelight UVO cleaner to generate OH groups at the surface. The SAM films were derived from DETA and OTMS. The DETA was vapour deposited onto the UV-ozone treated

Etching of DETA SAM resulting from atomic oxygen exposures

Fig. 2 displays the overall atomic concentration for the DETA SAM on the silicon substrate as measured by XPS throughout the experiment. Unlike conventional quantification methods which assume a homogenous surface, we assume a layered one where the Si/SiO2 substrate is attenuated by the SAM layer. It can be seen that as the carbon and nitrogen signals decrease the silicon and oxygen signals increase. This is interpreted as evidence of the removal of the SAM which exposes more of the underlying

Conclusion

Overall, due to excellent control of atomic oxygen exposures the stepwise etching of a SAM is observed. Upon loading both the DETA and OTMS SAMs, the photoemission spectra are essentially as expected, given their chemical composition. In the DETA SAM there is evidence of existing oxygen bonds in the photoemission spectra of as-received sample which has been attributed to atmospheric exposure. As part of the etching process for the DETA, the atomic oxygen appears to strip the terminal groups and

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to gratefully acknowledge that this project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No. 888163. The authors would also like to gratefully acknowledge financial support from the Science Foundation Ireland Principal Investigator programme under Grant No. 13/IA/1955. This work was carried out within the IMEC Industrial Affiliation Programme on Advanced Interconnects (IIAP).

References (58)

  • J. Yarbrough et al.

    Next generation nanopatterning using small molecule inhibitors for area-selective atomic layer deposition

    J. Vac. Sci. Technol. A.

    (2021)
  • A.J.M. Mackus et al.

    From the bottom-up: toward area-selective atomic layer deposition with high selectivity †

    Chem. Mater.

    (2019)
  • M. Fang et al.

    Area-selective atomic layer deposition: conformal coating, subnanometer thickness control, and smart positioning

    ACS Nano.

    (2015)
  • E. Färm et al.

    Self-assembled octadecyltrimethoxysilane monolayers enabling selective-area atomic layer deposition of iridium

    Chem. Vap. Depos.

    (2006)
  • X. Jiang et al.

    Area-selective ALD with soft lithographic methods: using self-assembled monolayers to direct film deposition

    J. Phys. Chem. C.

    (2009)
  • J.R. Avila et al.

    Real-time observation of atomic layer deposition inhibition: metal oxide growth on self-assembled alkanethiols

    ACS Appl. Mater. Interfaces.

    (2014)
  • D. Bobb-Semple et al.

    Area-selective atomic layer deposition assisted by self-assembled monolayers: a comparison of Cu Co, W, and Ru

    Chem. Mater.

    (2019)
  • R. Wojtecki et al.

    Fifteen nanometer resolved patterns in selective area atomic layer deposition - defectivity reduction by monolayer design

    ACS Appl. Mater. Interfaces.

    (2018)
  • L.i. Zheng et al.

    Area-selective atomic layer deposition of TiN using trimethoxy(octadecyl)silane as a passivation layer

    Langmuir.

    (2020)
  • R. Wojtecki et al.

    Additive lithography−organic monolayer patterning coupled with an area-selective deposition

    ACS Appl. Mater. Interfaces.

    (2021)
  • T.-L. Liu et al.

    Area-selective atomic layer deposition on chemically similar materials: achieving selectivity on oxide/oxide patterns

    Chem. Mater.

    (2021)
  • S. Herregods, B.T. Chan, I.N. Hetel, S. Armini, L. View, N. View, B.T. Chan, Vapour phase self-assembled monolayers for...
  • F.S. Minaye Hashemi et al.

    Selective deposition of dielectrics: limits and advantages of alkanethiol blocking agents on metal-dielectric patterns

    ACS Appl. Mater. Interfaces.

    (2016)
  • F.S.M. Hashemi et al.

    A new resist for area selective atomic and molecular layer deposition on metal-dielectric patterns

    J. Phys. Chem. C.

    (2014)
  • F.S.M. Hashemi et al.

    Sequential regeneration of self-assembled monolayers for highly selective atomic layer deposition

    Adv. Mater. Interfaces.

    (2016)
  • F.S. Minaye Hashemi et al.

    Self-correcting process for high quality patterning by atomic layer deposition

    ACS Nano.

    (2015)
  • D.J. Kim et al.

    A simple method for the removal of thiols on gold surfaces using an NH4OH-H2O2-H2O solution

    Scanning.

    (2008)
  • W. Lee et al.

    Oxidative removal of self-assembled monolayers for selective atomic layer deposition

    ECS Trans.

    (2008)
  • C. Prasittichai et al.

    Improving area-selective molecular layer deposition by selective SAM removal

    ACS Appl. Mater. Interfaces.

    (2014)
  • View full text