Guide: Viscoelastic Transfer of 2D Material Using PDMS – Part 1
|Written by Dr Evgeny Alexeev, a Post-doctoral Research Associate from Alexander Tartakovskii's 2D Materials group at the University of Sheffield in collaboration with Ossila.|
Viscoelastic transfer using polydimethylsiloxane (PDMS) stamps is one of the methods used for the deterministic placement of 2D materials and the fabrication of van der Waals heterostructures . It relies on the viscoelasticity of PDMS, which behaves as an elastic solid on short time scales, but as a viscous fluid on long time scales. As a result, PDMS films demonstrate kinetically controlled adhesion: the faster an object is pulled off the surface of the film, the stronger adhesion force it will experience. Faster peel-off rates therefore can be used to deposit 2D material onto PDMS, and slower rates - to release it onto a target substrate.
PDMS-based transfer offers a fast and convenient way of depositing 2D materials onto virtually any substrate and, unlike other methods, does not require the use of wet chemistry. One significant limitation of this technique are polymer residues left on the surface of the material after the transfer. These residues are generally inert and optically clear, and their presence might be beneficial for the optical properties of 2D crystals, acting as an encapsulation layer that protects the material from the surrounding environment. They can, however, prevent efficient interaction between different layers, and van der Waals heterostructures fabricated using this method usually require additional processing steps, such as thermal annealing or AFM ironing  to improve the quality of the interlayer interface. This technique is therefore best suited for the fabrication of individual monolayers and their integration into existing structures, such as pre-pattern contacts, photonic and plasmonic structures.
Preparation of the PDMS stamp
PDMS is a commercially available rubber, commonly used as protective coating for solar cells and light-emitting diodes. In this guide, we will be using pre-packaged films available from Gel-Pak (product code PF-3-X4), but two-component elastomer kits (such as Sylgard 184 from Dow Corning) can also be used. For viscoelastic transfer, a thin PDMS film (<1mm) is deposited onto a rigid transparent substrate, such as microscope glass slide, which acts as a support for mechanical exfoliation and transfer.
- Prior to deposition, glass slides need to be cleaned to remove any dust or organic residues that can interfere with stamp adhesion and lead to the formation of bubbles. Use nitrogen or a compressed air gun to blow dust off the glass substrates, then put them into UV Ozone Cleaner for 5 minutes. Alternatively, sonicate the substrates in a beaker of isopropyl alcohol (IPA) for 5 minutes, then blow dry with a nitrogen gun. If you don’t have access to an ozone cleaner or sonicator, cleaning the substrates with IPA wipes should give you sufficiently good results.
- Cut PDMS film into 5-10 mm squares. The PDMS film from Gel-Pak comes sandwiched between two protective films and can be safely cut using normal scissors.
- Separate the top two layers (PDMS and cover sheet) from the substrate using tweezers. Hold the PDMS by the corner of the stamp to minimise contamination.
- Carefully apply the film to the glass slide, avoiding the formation of bubbles. Then remove the protective cover sheet.
- If there are any wrinkles or bubbles formed, peel the film off and re-apply it to the substrate. Put one corner down first, then, slightly bending the rest of the stamp upwards, let it slowly roll onto the substrate.
Mechanical exfoliation is the method originally developed by Andre Geim and Konstantin Novoselov for the isolation of monolayer graphene in 2004 and has been successfully applied to other layered materials. Despite low yield and relatively small size of the single-layer flakes produced by this technique, it is still widely used in research due to its accessibility and excellent material quality of the produced samples. Single, and few-layer flakes are separated from bulk crystal and deposited onto a substrate using a piece of adhesive tape. We use high-quality tape (BT-150E-CM, Nitto) designed to minimise tape residues, but normal office tape, such as Magic Tape from Scotch, could also be used.
- Cut the tape into small pieces: they should be slightly larger than the PDMS stamp. Leaving some additional space on one side will make them easier to handle.
- Put the bulk crystal on one piece of tape and cover it with the second piece. Apply gentle pressure with flat-tipped tweezers to ensure good contact between the tape and the surface of the crystal.
- Peel the two pieces of tape apart, adjust their position slightly (so the bulk crystal overlaps with a clean piece of tape) and then bring them together again.
- Repeat the process, covering the surface of the tape with thin material. More peels will result in higher surface coverage and overall thinner flakes, but will cause the crystal to break into smaller part, resulting in a smaller size of the monolayer flakes.
- Additional pieces of tape can be created by transferring some of the prepared material onto a fresh piece of tape.
- After covering a large portion of the tape surface, and if the bulk crystal is still structurally sound, it can be removed from the tape with tweezers. Alternatively, the piece of tape with bulk crystal can be stored and used later for another exfoliation.
- Put one of the pieces of prepared tape on top of the PDMS stamp and apply gentle pressure with a cotton bud. The goal is to ensure good contact between the tape and PDMS, but applying too much pressure can lead to the formation of wrinkles and cracks in the material.
- Peel the tape off. Depending on the strength of the adhesion of the stamp to the glass substrate, you might need to hold the stamp down at a corner with tweezers.
- The rate of the peel defines the amount of material transferred onto PDMS: a faster peel rate will result in more material being transferred, but might also increase the average thickness of the flakes.
- Repeat this process with other pieces of tape and PDMS stamps that you have prepared.
Flake search and thickness identification
Mechanical exfoliation produces flakes of various sizes and thicknesses; atomically thin flakes are a minority among much thicker crystals. These flakes need to be located and identified using a reflected-light microscope – a process commonly known as flake search. Mono, and few-layer flakes can be distinguished by their translucent appearance and low optical contrast. Their visibility will depend on the material, substrate it is deposited on, and the optical system used. Materials with higher absorption in visible range, such as transition metal dichalcogenides, will be more visible than materials with weaker absorption such as graphene or hexagonal boron nitride. Certain types of substrates, such as silicon wafers covered with 90 nm or 290 nm thick silicon dioxide layer, can improve the flake visibility. Once a suspected monolayer flake is located, its thickness can be estimated using optical contrast measurements and confirmed using Raman or photoluminescence spectroscopy.
- Inspect the substrate using a reflected-light optical microscope to look for the most translucent flakes. Higher-magnification lenses with a shallow depth of field usually provide a better optical contrast for PDMS stamps.
- Once a possible monolayer flake is located, save its digital image (making sure that is properly exposed). Take additional images at lower magnifications and note the approximate position of the flake on the PDMS stamp to help you find it later.
- Open the acquired high-magnification image with ImageJ (https://imagej.nih.gov/ij/) or Fiji (https://fiji.sc/).
- Select an area of the suspected monolayer using one of the selection tool and measure the mean intensity of the reflected light by pressing M on the keyboard. The data will be displayed in the ‘Results’ window. The type of measurement displayed in the window can be changed by right-clicking on the column title and selecting ‘Set Measurements’.
- Move the selection region onto the substrate and repeat the measurement. Try to keep the distance between the measurement points as small as possible to avoid the influence of the non-uniform illumination.
- Calculate the optical contrast as:
The optical contrast of layered 2D materials changes linearly with increasing number of layers. The value of the contrast can be positive or negative, depending on the type of the substrate used, and might also vary depending on the measurement conditions. However it should be close to the minimum difference between the adjacent thicknesses.
- Monolayer thickness can be confirmed using optical spectroscopy. For graphene, the properties of the Raman 2D peak located at 2700 cm-1 are commonly used to identify single-layer flakes. It has a symmetric lorentzian shape and FWHM ∼ 25-30 cm-1 in monolayer, but broadens substantially in bilayer (FWHM ∼ 50 cm-1) and requires four lorentzians to accurately reproduce its shape .
- Similarly, for hBN the frequency and intensity of the G peak at ~1360 cm-1 can be used , although a few-layer hBN will likely be completely invisible on PDMS stamps due to the low optical contrast.
- The most convenient way to identify TMD thickness is using PL imaging or spectroscopy. Group-VI TMD have direct band gap only in their monolayer form, so the intensity and shape of the PL peak can be used for thickness identification . Monolayer TMDs will show a bright and spectrally narrow emission peak. In contrast, photoluminescence signal in bilayers will be much weaker and have a significantly broader linewidth, showing emission originating from both momentum-direct and indirect transitions. Further increase in thickness will lead to almost complete loss of the PL signal.
Novoselov, K. S., Mishchenko, A., Carvalho, A. & Castro Neto, A. H. 2D materials and van der Waals heterostructures. Science (80-. ). 353, aac9439 (2016) – a comprehensive review of 2D materials and van der Waals heterostructures
Huang, Y. et al. Reliable Exfoliation of Large-Area High-Quality Flakes of Graphene and Other Two-Dimensional Materials. ACS Nano 9, 10612–10620 (2015) – a paper detailing mechanical exfoliation of 2D materials onto SiO2 /Si substrate.
Alexeev, E. M. et al. Imaging of Interlayer Coupling in van der Waals Heterostructures Using a Bright-Field Optical Microscope. Nano Lett. 17, 5342–5349 (2017) – a paper from our group showing how a bright-field microscope can be used for photoluminescence imaging of monolayer TMDs and their heterostructures
Jain, A. et al. Minimizing residues and strain in 2D materials transferred from PDMS. Nanotechnology 29, 265203 (2018) and Schwartz, J. J. et al. Chemical Identification of Interlayer Contaminants within van der Waals Heterostructures. ACS Appl. Mater. Interfaces acsami.9b06594 (2019). doi:10.1021/acsami.9b06594 – two papers describing methods of reducing surface contamination in PDMS-transferred samples.