Absorption of Water Molecules on a TiO2 Surface and the Water Gas Rxn Have Been Visualized.

[NNadir]

The paper I'll discuss in this post is this one: Visualizing H2O molecules reacting at TiO2 active sites with transmission electron microscopy (Wentao Yuan1,*, Beien Zhu2,3,*, Xiao-Yan Li2,4,*, Thomas W. Hansen5, Yang Ou1, Ke Fang1, Hangsheng Yang1, Ze Zhang1,†, Jakob B. Wagner5,†, Yi Gao2,3,†, Yong Wang1,† Science 2020:
Vol. 367, Issue 6476, pp. 428-430)

The key to reversing the climate change now underway at an accelerating rate is almost certainly thermochemical carbon dioxide splitting into carbon monoxide and oxygen, using transition metal oxides or cerium oxide and nuclear heat

. Effectively access to carbon monoxide is equivalent to access to hydrogen using the water gas reaction CO + H2O <-> H2O + CO2. The overwhelming majority of the hydrogen produced industrially on this planet (>98%) for captive use (the only safe and appropriate use) is produced using this reaction, the CO being produced from the partial oxidation of dangerous natural gas.

From my perspective, a key material to make a sustainable future is the relatively common element titanium, in both its metallic form - which is now accessible at reduced costs using the FFC Cambridge process - and its oxide, which turns out to be a very valuable catalyst for splitting water using CO. (It can also photochemically split water, but decades of research have failed to make this process anywhere near being capable of industrialization.)

This paper, out of China, does something quite remarkable, which is to obtain images of water molecules on the surface of titanium oxide.

From the text:

Imaging at the atomic scale with transmission electron microscopy (TEM) has benefited from the developments of aberration correctors and in situ equipment (1–8). For studies of heterogeneous catalysts, these developments, along with approaches that allow gases and even liquids to contact samples [known as environmental TEM (ETEM)], have enabled imaging of single molecules and atoms adsorbed on a catalyst surface (9–14). However, the direct visualization of gas molecules reacting at catalytic sites is generally difficult to achieve with TEM. Normally, the molecules that adsorb and react dynamically do not offer sufficient contrast for TEM identification. We now show that this obstacle can be overcome by taking advantage of the highly ordered four-coordinated Ti (Ti4c) rows (termed “active rows,” owing to their lower coordination) on the anatase TiO2 (1×4)-(001) surface [i.e., a TiO2(001) surface with (1 × 4) reconstruction] to facilitate enhanced contrast of adsorbing molecules along the row direction and allow real-time monitoring of H2O species dissociating and reacting on the catalyst surface.

The atomic structure of the TiO2 (1×4)-(001) surface has been characterized by both aberration-corrected ETEM and scanning transmission electron microscopy (STEM) images. The bulk-truncated (1×1)-(001) surface usually reconstructs to a (1×4)-(001) surface (Fig. 1, A to C) by periodically replacing the surface oxygen rows (along the [010] direction) with TiO3 ridges every four unit cells along the TiO2[100] direction (15–17). As a result, protruded Ti4c rows are periodically exposed on the surface and show distinct contrast, so the subtle changes occurring in reactions could be detected by means of ETEM observation without contrast overlap. The ordered Ti4c active rows could provide sufficient contrast for direct ETEM visualization of water if the molecules adsorbed in ordered arrays...

Figure 1:



The caption:

ig. 1 Dynamic atomic structural evolution of the (1×4) reconstructed TiO2(001) surface in a water vapor environment.
(A) High-angle annual dark-field–STEM image of the (1×4)-(001) surface, viewed from the [010] direction. The image was acquired at 700°C in vacuum (TEM column pressure: ~10?7 mbar). (B) ADM reconstruction models of the (1×4)-(001) surface (Ti, gray; O, red). (C) Atomic models of a Ti4c row. (D to G) Aberration-corrected in situ ETEM images show the same area of TiO2(001) surface at 700°C under oxygen [(D), 0.001 mbar] and water vapor [(E), 0.01 mbar; (F), 1 mbar; (G), 2.5 mbar] conditions. Scale bar, 1 nm. (H to J) Another case shows the reversible structural transition induced by a change in the gas environment at 700°C from oxygen [(H), 0.001 mbar] to water vapor [(I), 3 mbar] and then reversion to oxygen [(J), 0.001 mbar]. Scale bar, 2 nm.

The authors restructured the common mineral form of TiO2 (rutile) by heating it in oxygen, evacuating the oxygen, and then exposing the reconstructed mineral to steam.

...We synthesized TiO2 nanocrystals with exposed {001} facets by a hydrothermal route (see supplementary materials) (18, 19). The nanocrystals were heated in oxygen in situ (~10?3 mbar) at 500° to 700°C to trigger the reconstruction. The reconstructed structures remained stable in this temperature range, in accord with recent ETEM studies (15, 16, 20). During the ETEM experiments, we used a constant electron beam dose with a small value (<1 A/cm2), and no appreciable irradiation damage was observed on the TiO2 surface (21). After heating at 700°C for ~10 min, the reconstructed TiO2 (1×4)-(001) surface of an ad-molecule (ADM) configuration was obtained, as confirmed by the ETEM image (Fig. 1D), in which the protruding black dots represent the Ti4c rows. The ADM structure did not change appreciably after ~16 min of intermittent TEM observation.

The O2 gas was then evacuated, and H2O vapor (fig. S1) was introduced at the same temperature. When the H2O pressure was raised to 1 mbar, two additional small protrusions were observed at the top of the Ti4c rows (Fig. 1F). This twin-protrusion structure became more resolved for a H2O pressure of 2.5 mbar, owing to a higher water surface coverage (Fig. 1G and movie S1)...

Figure 2 of the paper:



The caption:

Fig. 2 The twin-protrusion configuration of adsorbed water.
(A) In situ FTIR spectra of the hydroxyl region for TiO2 in the presence of water vapor (5 mbar; 500°C) and vacuum (10?6 mbar; 500°C). The inset shows results of a theoretical simulation. (B to D) Atomic structure of the adsorbed H2O species on the TiO3 rows, as verified by theoretical calculations, viewed from the [010] direction (B), the [100] direction (C), and the [00-1] direction (D) (gray, Ti; red, O; cyan, H).

Visualizing the water gas reaction:

Because TiO2 can catalyze the water–gas shift reaction (H2O + CO ? H2 + CO2) at elevated temperatures (28, 29), we studied this reaction by introducing CO into the ETEM column. The gas environment was changed from pure water vapor (2.5 mbar) to a mixed gas environment (CO and H2O vapor in a 1:1 ratio; pressure: 5 mbar). Under these conditions, the twin-protrusion structure became unstable (Fig. 3A and movie S2). Its contrast changed dynamically: Most of the time it was blurred, but it would occasionally clear (Fig. 3B), with no substantial contrast change observed in TiO2 bulk and in other surface areas. For example, in one case the twin protrusion was clearly seen initially [Fig. 3B, (1)], almost disappeared after 2.2 s [Fig. 3B, (2)], and then reappeared at 4 s [Fig. 3B, (3)]. The disappearance and reappearance occurred again at 5.8 s [Fig. 3B, (4)] and 7.8 s [Fig. 3B, (5)], respectively. The contrast change of the twin protrusions was also evidenced by the intensity profiles across the protruding row (Fig. 3C). Similar cases are shown in fig. S11 and movie S3. In a pure water vapor environment, the twin protrusions did not display such contrast changes (fig. S12 and movie S1), hence ruling out electron beam effects for the disappearances and reappearances.

Figure 3:



The caption:

Fig. 3 Dynamic structural evolution of the (1 × 4)-(001) surface in the water–gas shift reaction.
(A) Sequential ETEM images acquired in the mixed gas environment (1:1 ratio of CO and H2O vapor; gas pressure: 5 mbar; temperature: 700°C), viewed from the [010] direction. Scale bar, 2 nm. (B) Enlarged ETEM images show the dynamic structural evolution of the Ti row outlined by the dotted rectangle in (A). Scale bar, 0.5 nm. (C) Intensity profiles along the lines crossed the Ti rows of (B). Blue arrows denote intensity valleys corresponding to the twin protrusions. a.u., arbitrary units.

My son has been working quite a bit with stuff on the atomic scale, both in imaging. (This summer he had a chance to work around neutron scattering experiments.) I'll have to share this cool paper with him.

Have a nice weekend.

[5X]

1. I always thought the molecular models that appear in textbooks were just a

representation of molecular structure, not an actual depiction, but guess I was wrong.

[NNadir]

2. The cartoons are somewhat clearer than the images. We've been able to "see" at an atomic...

...scale for quite some time. When it first was achieved it was front page news, if I recall correctly it involved xenon atoms arranged on a platinum surface in such a way as to spell out "IBM."

The ease with which atoms can be visualized depends highly on what kind of atoms they are. Heavier atoms are easier to see than lighter atoms.

When I toured the materials science building with my son when we were checking out the university to which he ultimately went - they have a lot of electron microscopes there - I was amazed at all the crystal structures they could pull up there. They are able to add tags saying which element each atom is.

This paper is striking however because it details light atoms, hydrogen, oxygen, and carbon.