High-energy resolution X-ray absorption and emissionspectroscopy reveals insight into unique selectivity ofLa-based nanoparticles for CO2Ofer Hirscha, Kristina O. Kvashninab,c, Li Luoa, Martin J. Süessd, Pieter Glatzelb, and Dorota Kozieja,1

aDepartment of Materials, Laboratory for Multifunctional Materials, Eidgenössische Technische Hochschule Zürich, 8093 Zurich, Switzerland; bEuropeanSynchrotron Radiation Facility, 38000 Grenoble, France; cHelmholtz Zentrum Dresden-Rossendorf, Institute of Resource Ecology, 01314 Dresden, Germany;and dDepartment of Physics, Laboratory for Quantum Optoelectronics, Eidgenössische Technische Hochschule Zürich, 8093 Zurich, Switzerland

Edited by Jeffrey T. Miller, Argonne National Laboratory, Argonne, IL, and accepted by the Editorial Board November 19, 2015 (received for reviewAugust 14, 2015)

The lanthanum-based materials, due to their layered structure andf-electron configuration, are relevant for electrochemical applica-tion. Particularly, La2O2CO3 shows a prominent chemoresistive re-sponse to CO2. However, surprisingly less is known about its atomicand electronic structure and electrochemically significant sites andtherefore, its structure–functions relationships have yet to beestablished. Here we determine the position of the different con-stituents within the unit cell of monoclinic La2O2CO3 and use thisinformation to interpret in situ high-energy resolution fluorescence-detected (HERFD) X-ray absorption near-edge structure (XANES) andvalence-to-core X-ray emission spectroscopy (vtc XES). Comparedwith La(OH)3 or previously known hexagonal La2O2CO3 structures,La in the monoclinic unit cell has a much lower number of neigh-boring oxygen atoms, which is manifested in the whiteline broad-ening in XANES spectra. Such a superior sensitivity to subtle changesis given by HERFD method, which is essential for in situ studying ofthe interaction with CO2. Here, we study La2O2CO3-based sensors inreal operando conditions at 250 °C in the presence of oxygen andwater vapors. We identify that the distribution of unoccupied Lad-states and occupied O p- and La d-states changes during CO2 che-moresistive sensing of La2O2CO3. The correlation between these spec-troscopic findings with electrical resistance measurements leadsto a more comprehensive understanding of the selective adsorp-tion at La site and may enable the design of new materials for CO2

electrochemical applications.

lanthanum oxycarbonate | HERFD XAS | valence-to-core XES | structure |CO2 sensing

CO2 has become a challenge for our society and we have todevelop new materials for its photo/electrocatalysis, chemo-

resistive sensing, and storage (1–8). Particularly, for the variety ofelectrochemical applications the selective interaction of CO2 andcharge transfer with solids is in the foreground. At the same time,the interaction of CO2 with solids in the electrochemical cell orsensing device is rather complex, thus it remains challenging toexperimentally identify the key elements determining their se-lectivity and efficiency. X-ray absorption spectroscopy (XAS) andX-ray emission spectroscopy (XES) provide complementary in-formation on the electronic structure of materials (9, 10) and onthe orbitals participating in the interaction with absorbing molecules(11). High-energy resolution fluorescence-detected (HERFD) XASprobes unoccupied states with a spectral resolution higher thanregular XAS. Furthermore, with the same experimental setupXES can be measured, which allows one to probe the occupiedstates within the valence band (12). In situ HERFD XAS or XESexperiments have been previously carried out to study the cata-lytic reaction at the surface of noble metals (11, 13–16), zeolites(17), and metal organic frameworks (18). Thus far, no such in situexperiments have been performed to directly track the changes ofthe electronic structure of a solid and its electrochemical activitytoward CO2. The rare-earth–based materials like perovskites and

oxycarbonates, owing to their unique f-electron configuration ofLn (Ln = rare earth) and layered crystal structure, emerge as themost interesting for future photo- and electrochemical applica-tions (3–8). Among rare-earth oxycarbonates (19, 20), particularlylanthanum strongly responds to CO2 and shows up to 16-foldconductivity changes, not seen before for any metal oxides (21).This is very surprising because a direct injection of an electron intoCO2 molecule requires the activation energy of nearly 2 eV (22).To assess the origins of the unique CO2 sensitivity of rare-earthoxycarbonate, it is essential to study in situ the interplay betweenthe changes of the electronic structure of La-based nanoparticlesupon CO2 adsorption and changes of the macroscopic conduc-tivity of a device.Here, to elucidate the underlying mechanism we first determine

the structure and atomic positions of the lanthanum oxycarbonate.Using HERFDXAS and valence-to-core (vtc) XES results, we gaininformation about the electronic structure and band gap. Moreover,we combine in situ HERFD XAS and XES measurements withsensing performance tests to obtain the structure–function re-lationship. Finally, with all of the obtained information we discussa mechanism of CO2 adsorption on the La2O2CO3 surface.

Results and DiscussionThe synthesis and characterization of La2O2CO3 nanoparticlesare described in SI Experimental Methods, Synthesis and Char-acterization, including Figs. S1–S3. The crystallite size is between11 and 14 nm as shown in the high resolution transmissionelectron microscopy (HRTEM) images in Fig. 1 A and B and

Significance

CO2 has become a challenge for our society and we have todevelop new materials for its photo-/electrocatalysis, chemo-resistive sensing, and storage. Particularly, for the variety ofelectrochemical applications the selective interaction of CO2

and charge transfer with solids is in the foreground, but theirorigins are poorly understood. Our story will undoubtedlyshowcase how to access the key information, which is relevantfor electrochemical application from in situ X-ray absorptionspectroscopy/X-ray emission spectroscopy studies.

Author contributions: O.H. and D.K. designed research; O.H., M.J.S., P.G., and D.K. per-formed research; O.H., K.O.K., L.L., M.J.S., P.G., and D.K. analyzed data; and O.H., K.O.K.,L.L., P.G., and D.K. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission. J.T.M. is a guest editor invited by the EditorialBoard.

Data deposition: The atomic coordinates and structure factors have been deposited in theCambridge Structural Database (CSD), Cambridge Crystallographic Data Centre, Cam-bridge CB2 1EZ, United Kingdom [CSD reference no. 430439 (La2CO5)].1To whom correspondence should be addressed. Email: dorota.koziej@mat.ethz.ch.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1516192113/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1516192113 PNAS | December 29, 2015 | vol. 112 | no. 52 | 15803–15808

APP

LIED

PHYS

ICAL

SCIENCE

S

Dow

nloa

ded

by g

uest

on

Mar

ch 2

6, 2

020

Fig. S1A. Three different polymorphs of rare-earth oxycarbonatesare known: a tetragonal type I, the monoclinic type Ia, andthe hexagonal type II. All these types share a layered structurewith alternating Ln2O2

2+ and CO32− layers (23, 24). In the

hexagonal type this layer consists of distorted LnO8 coordinatedrhombohedra (19). Whereas the local coordination around theLn atom of type II polymorph is known (25), the atomic posi-tions within the type Ia unit cell are not resolved yet. Here,we assign the powder X-ray diffraction (PXRD) peaks solely tothe Ia polymorph of La2O2CO3 as shown in Fig. 1C. TheLa2O2CO3 nanoparticles crystallize in the space group P21/c withthe refined lattice parameters a = 4.0756 Å, b = 13.4890 Å,c = 5.8034 Å, α = γ= 90°, β = 135.37°, as deduced from theRietveld analysis of the PXRD (26–35). The lattice parametersand the atomic positions of La, C, and O in the La2O2CO3 latticeand a detailed refinement procedure are given in SI ExperimentalMethods, Structure Determination, and Table S1. The analysis ofthe atomic positions within the structure helps us to identify localstructural commonalities and differences between the monoclinicand the hexagonal polymorph. The Ln2O2

2+ layer in the hexagonaltype II consists of distorted LnO8 coordinated rhombohedra (19),whereas the same layer in the type Ia polymorph contains pyramidal-oriented oxygen with a Ln atom in the middle as shown in Fig. 1D.The pyramids are pointing in positive or negative b direction andalternate along the c direction (Fig. S4). Within a distance (R) of 3 Åthe lanthanum atom has six oxygen neighbors in type Ia, whereas Lahas eight oxygen neighbors in type II. Both crystal structure and thecoordination of the metal cations and the oxygen anions are ingeneral known to have strong impact on the chemical reactivity ofrare-earth–based nanoparticles (36). However, reports correlating

the structure of La2O2CO3 with chemical reactivity are relativelysparse. For example, the crystal structure of intermediate La2O2CO3determines the efficiency of methane reforming to produce hydro-gen (37, 38). Additionally, it was previously shown that aLa2O2CO3-based sensor shows a higher sensor signal thanNd2O2CO3 (19, 21). No further information was given about theorigin of these differences, but remarkably Nd oxycarbonatecrystallizes in the hexagonal type II, whereas La oxycarbonatecrystallizes in monoclinic type Ia structure. The formation ofLn2O2CO3 from Ln(OH)3 is a sequential process involvingcompositional and structural modifications. The heating rate,temperature, and CO2 concentration determine the free energyof transformations and thus the occurrence of the individualpolymorphs.We measured HERFD XAS (Fig. 2A) to obtain information

about the unoccupied states and vtc XES to retrieve informationabout the occupied states (Fig. 2B) of La2O2CO3. The largelifetime broadening at the 2p level [∼3.4 eV (39)] renders theXAS technique at the La L3 edge rather insensitive to the elec-tronic structure. Especially, the preedge structure of the La L3XAS spectrum is very difficult to resolve with conventional XASexperiment. During the HERFDmeasurements, the X-ray emissionspectrometer is tuned to the maximum of the Lα1 (3d5/2–2p3/2)transition and the absorption is recorded by monitoring the maxi-mum of the Lα1 intensity as a function of the incident energy.The advantage of such a setup is that the width of the spectralfeatures is no longer limited by the 2p3/2 core-hole lifetime but bythe sharper 3d5/2 core-hole width in the final state ∼0.7 eV (40),which is on the same order of magnitude as the experimentalbroadening. To test the sensitivity of the HERFD XAS methodfor the oxygen coordination of La, we compare the spectrum oflanthanum oxycarbonates (six oxygen neighbors) with lanthanumhydroxide (nine oxygen neighbors), as shown in Fig. 2A. The ex-perimental data show the maximum absorption intensity locatedat the same energy, which corresponds to a La3+ ion as it is expectedfrom the stoichiometric formula. There are two important dis-crepancies between the spectra of these two compounds; La2O2CO3exhibits a wider whiteline, whereas La(OH)3 exhibits a higherwhiteline intensity. These observations imply that the charge dis-tribution is more localized around the La ion in the hydroxidethan in oxycarbonate, whereas the formal charge stays three plus.The spectra at energies above the absorption edge reveal morestructural features for the hydroxide than for the oxycarbonate.To understand the differences in the experimental data, we calcu-late XAS spectra with the FEFF program package as shown inFig. 2A, dashed lines. The shape of the whiteline as well as thepreedge structure are reproduced for both compounds, whereasthe postedge features are not well modeled in the calculations.Furthermore, the calculation reproduces the differences in thewhiteline width and intensity well. The atomic orbital angular mo-mentum projected density of states (DOS) of both compounds re-veal the electronic states that give rise to the preedge and thewhiteline excitations and are plotted in Fig. 2 C and D for La(OH)3and La2O2CO3, respectively. The width of the whiteline andtherefore the width of the unoccupied d-states reflects the numberof oxygen neighbors within the first shell around the La absorber(41, 42). The higher the coordination number of La, the narroweris the whiteline. Also the crystal structure may induce similarchanges of the spectral features. Fig. 2 E and F shows atomswithin a sphere of 3 Å for La(OH)3 and La2O2CO3, respectively.We count nine nearest oxygen neighbors for La(OH)3 whereasLa in La2O2CO3 has only six nearest oxygen neighbors. Bandstructure calculations of the electronic states of La2O3 with itsseven La–O pairs within the sphere of 3 Å show also a wide dis-tribution of empty d-states (43, 44). In the spectra of La(OH)3and La2O2CO3 we observe a preedge feature at 6.7 eV below themaximum of the La L3 absorption edge, which is generally attrib-uted to the mixed dipole and quadrupole transitions (40, 45, 46).

Fig. 1. Structure and morphology of the La2O2CO3 nanoparticles. (A and B)HRTEM images of La2O2CO3 nanoparticles at different magnifications.(B, Inset) Fourier transform of the region highlighted in red including thezonal axis and indexed reflections. (C) Recorded PXRD pattern of monoclinicLa2O2CO3 (black line), the calculated (red line), and the difference (blue line)resulting from the Rietveld analysis. Reference patterns of the three poly-morphs are shown [monoclinic, hexagonal (Inorganic Crystal Structure Da-tabase [ICSD]: 37–0804), tetragonal (ICSD: 23–0320)]. (D) A model of therefined La2O2CO3. The lanthanum atoms are cyan, oxygen red, and carbonyellow. The polyhedrons denote the local La surrounded by pyramidal-oriented O atoms.

15804 | www.pnas.org/cgi/doi/10.1073/pnas.1516192113 Hirsch et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

6, 2

020

Only systems with inversion symmetry at the absorbing atomcan have a preedge structure of pure quadrupole contribution,but if the inversion symmetry is broken the quadrupole and di-pole transitions contribute to the preedge intensity (47). Thecompounds are not having an inversion center at La atom andthus we observe mixed d- and f-states in the preedge structure.The analysis of the preedge DOS reveals a small contribution ofdipole transitions of d-state character and strong contribution ofquadrupole transition of f-state character. However, the intensityof quadrupole transitions is considerably smaller than the in-tensity of dipole transitions, thus the measured preedge featureis very small (45).

To obtain information about the occupied states we performresonantly excited XES measurements. We therefore tune theincident photon energy to the maximum of the absorption edgeand detect emission spectra with energy differences from 0 to−70 eV with respect to the incident energy and probe thereforethe highest occupied states. Fig. 2B shows the experimental XESspectra of La(OH)3 and La2O2CO3. The main features for bothcompounds are very similar and located at energies at −10 eVand at −40 eV. Interestingly, even at lower transfer energies, 50and 60 eV below the whiteline maximum some features aredetectable, which show some small differences in relative in-tensity between the hydroxide and the oxycarbonate. The mostpronounced differences are visible in the features which corre-spond to the highest occupied levels. Furthermore, the shapeof the structure at −10 eV also differs. For spectral features atenergies close to the Fermi level at −10 eV we assume, basedon earlier calculations of La(OH)3 and La2O3, a mixing of Lad-states and O p-states (43, 44, 48–50). We observe a feature at−25 eV, which we can assign to a mixing of La p- and O s-stateson the basis of calculations on La2Ti2O7 (51). The feature at−40 eV was previously observed in X-ray photoelectron spec-troscopy spectra of La(OH)3 and La2O3, and was assigned to La5s states. This feature is showing an additional shoulder, whichis assigned to a difference in charge transfer from the ligand inLa(OH)3 and La2O3. In XES spectra of La(OH)3 and La2O2CO3these differences are also clearly visible, as shown in Fig. 2B(Inset). Within La(OH)3 and La2O2CO3 the formal valence oflanthanum ions is La3+ and the d-states are empty in the groundstate. In both cases occupied La d-states exist and they form,together with O p-states, the valence band. The La d-states areonly partly occupied and the empty fraction of the La d-statesforms the conduction band. The valence and conduction bandare separated by an electronic band gap Eg. Combining XAS andXES provides a gap in the d-DOS integrated over all directionsof momentum transfer and over a range of modulus over mo-mentum transfer that is given by the experimental geometry (52–54). Here we determine electronic d-DOS band gap energies of5.1 and 3.7 eV for La(OH)3 and La2O2CO3 nanoparticles, re-spectively. These values are lower with respect to the opticalband gap values derived from UV-Vis spectroscopy with theassumption of an indirect band gap; 5.4–5.6 eV and 4.35 eV formicrometer-large La(OH)3 and La2O2CO3 particles, respectively(55, 56). However, the direct comparison is not straightforwardbecause UV-Vis probes the DOS that is projected according toselection rules for the optical transition. At this wavelength themomentum of a photon is very small compared with the wavevector of the electrons and thus most of the UV-Vis signal arisesfrom direct transitions between valence and conduction band,whereas the indirect transitions are less pronounced.The chemoresistive sensing principle involves interaction of

CO2 molecules with sensing materials, which then leads to con-ductivity changes. In real applications, the interaction is not anisolated CO2 reaction but it is altered by the presence of oxygen andhumidity. For example, in situ diffuse reflectance infrared Fourier-transformed spectroscopy studies revealed for Nd2O2CO3-basedsensor a correlation between the amount of surface carbonatesgroups and adsorbed hydroxyl groups; the higher the amount ofadsorbed carbonates the lower the amount of water-relatedspecies on the surface. However, it was suggested that the twogases CO2 and H2O do not compete for the same adsorption sites(19). A similar reaction mechanism has been postulated forLa2O2CO3-based sensors, but so far the sites for CO2 adsorptionhave not been identified (21). Therefore, we are particularly in-terested in determining the reaction which induces the chargetransfer between La2O2CO3 and CO2 in humid air, and as a resultdecreases the overall conductivity of the La2O2CO3-based sensor.Before we discuss the in situ spectroscopic data we briefly sum-marize the sensing characteristic of La2O2CO3-based sensors and

Fig. 2. Ex situ HERFD XAS and vtc XES experiments along with FEFF cal-culations and coordination of La atom for La(OH)3 and La2O2CO3. (A) Acomparison of the experimental HERFD XAS spectra (solid lines) with theFEFF-calculated spectra (dashed lines): La(OH)3 (black) and La2O2CO3 (red).(B) Comparison of measured vtc XES spectra of La(OH)3 and La2O2CO3. Weobserve differences at the feature at −10 eV and at the features at −50 eVand −60 eV, which are magnified in the inset. We assume the first feature iscaused by La d- and O p-states, the second feature by La p- and O s-states,and the feature at −40 eV by La s-states. (C ) La(OH)3 and (D) La2O2CO3 DOSof the La absorber atom. The width of d-DOS width and therefore thewhiteline width relates to the amount of oxygen neighbors of the Laatom, which are shown in E and F as cluster of LaOx bonds in lanthanumhydroxide and oxycarbonate. The cutoff length of the sphere around theLa atom (R) was set to 3 Å. The lower coordinated La atom in La2O2CO3

exhibits a broader distribution of unoccupied d-states than the ninefoldcoordinated La(OH)3.

Hirsch et al. PNAS | December 29, 2015 | vol. 112 | no. 52 | 15805

APP

LIED

PHYS

ICAL

SCIENCE

S

Dow

nloa

ded

by g

uest

on

Mar

ch 2

6, 2

020

the impact the incident X-rays thereof. The resistance of theLa2O2CO3-based sensor decreases, and upon exposure to CO2 orCO increases as shown in Fig. 3A and Fig. S5 A and B. Moreover,we observe the positive cross-sensitivity to CO2 and humidity: Thebaseline resistance of La2O2CO3-based sensor at 250 °C is de-creasing with increasing humidity, and exposure of CO2 in humidconditions results in higher relative resistance changes than in dryair, as shown in Fig. S5C. We define the sensor signal accordingto the definition for p-type semiconductors as the resistance ofthe sensor upon exposure to reducing gas (RCO or RCO2) over thebaseline resistance of the sensor before exposure (R0) (57).For the in situ XAS and XES experiments we used condi-

tions resulting in the highest sensor signal, which is a pulse of

10,000 ppm of CO2 in synthetic air with 50% relative humidity (rh)and at operating temperature of 250 °C. The correspondingchanges in the sensor signal during in situ HERFD XAS mea-surements under air and CO2 pulse are shown in Fig. 3 B and C.At a given condition two sets of 15 spectra each were recorded.The labels “X-rays on/off” indicate that the shutter was open orclosed, respectively. The incident beam even at energies below theedge causes a rapid decrease of the sensor’s resistance of abouttwo orders of magnitude, which results in sensor signal of morethan 150 as shown in Fig. 3C. During an XAS scan the resistanceshows a dependence on the incident X-rays, which indicates thatelectrons are escaping from the La2O2CO3 layer as a result ofabsorption of an X-ray photon. At an energy corresponding to themaximum of whiteline intensity the sensor signal increases evenfurther to almost 400. The X-ray–induced changes of resistancehave been previously used to record the total electron yield-XASspectra (58). However, for the wide-band semiconductors andinsulating materials, the charging of the sample has to be experi-mentally or mathematically compensated by taking into accountthe effective penetration depths of secondary electrons as afunction of incident X-ray energy (58–61), and thus we will notdiscuss it any further. Instead, we analyze the impact of the in-cident X-rays on the reactivity of La2O2CO3 toward CO2. Weobserve that between the XAS scans, when the X-ray shutter isshortly closed, the resistance/sensor signal only partially recoversas shown in Fig. 3 B and C; but if enough time is allowed, it fullyrecovers to the initial value as shown in Fig. S6. In summary, theresistance of the sensor increases 16-fold upon exposure to10,000 ppm of CO2; the resistance of the sensor decreases almost400-fold upon irradiation with X-rays and depends on the energyof incident beam as shown in Fig. 3 B and C. The overall re-sistance of the sensor exposed to CO2 under X-ray irradiation isclearly higher than in air, as shown in Fig. S6. The direct quan-titative comparison of sensor signal toward CO2 with and withoutX-ray irradiation is not possible. We assume that the mechanismunderlying the chemical reactivity of La2O2CO3 toward CO2 isnot affected by incident X-rays but only the transduction of thechemical reaction into the electrical signal, and thus we canqualitatively compare resistivity changes as a function of gasenvironment under exposure to X-rays. Additionally, the sim-plified model for the resistance of the sensor at different ex-perimental conditions is given in Fig. 3D. The XAS and XESspectra of films at 250 °C show identical features as spectra ofpellets at room temperature as shown in Figs. 2 A and B and 4,

Fig. 3. Overview on the CO2 sensing performance of La2O2CO3-basedsensor with and without X-ray irradiation. (A) Sensing performance uponexposure to CO2 from 250 to 10,000 ppm in 50% rh at 250 °C withoutX-ray irradiation. (B) Comparison of the sensor signal upon exposure to10,000 ppm of CO2 without (blue) and with X-ray irradiation (black). Thesensor signal is defined as RCO2/Rair, where Rair and RCO2 are the resistance ofthe sensor in air and under CO2 exposure, respectively. Resistance underX-ray irradiation is energy dependent, thus we use as Rair the initial valuebefore the X-ray irradiation. During 0–3 h under air, we first measure twosets of 15 X-ray absorption near-edge structure (XANES) scans each. Betweenthe individual scans the fast X-rays shutter is closed (X-rays off) and open(X-rays on), which is visible as a spike in the sensor signal. Finally, we measureonly the resistance for 20 min. We repeat the same data acquisition protocolfor the measurements under CO2. (C) Comparison of the simultaneouslyrecorded XANES spectra (red) and the sensor signal (black) illustrates thedependence of resistance of the sensor on the incident energy. XANES andthe sensor signal measurements were measured with different temporalresolution. One measurement point of the sensor signal corresponds to 17points at the energy scale of XANES scan (1.7 eV). (D) The simplified modelfor the resistance of the sensor at different experimental conditions shown inC. We assume that at 250 °C the Debye length (LD) is smaller than theradius (r) of La2O2CO3 nanoparticles. Gases adsorbing at the surface create adepletion/accumulation layer with the thickness of LD. The resistance of thislayer (Rs) is different from the resistance of bulk of nanoparticles (Rb). Thecharge carriers have to overcome an energy barrier (eVs) to reach the neighboringnanoparticle. Dependent on the conditions the height of the barrier is changingas schematically shown (57); the escape depth of photoelectrons emitted from theabsorbing atom is larger than the size of nanoparticles (58), which means thatX-rays equally change the resistance at the surface as well as in the bulk ofLa2O2CO3 nanoparticles. The colors represent the measured resistance.

Fig. 4. In situ studies of structure–functions relationship. (A) The sensorsignal upon exposure for 3 h to 10,000 ppm of CO2. (B) The schematic of unitcell of La2O2CO3 illustrating the layered structure and the sixfold co-ordination of La atoms. La, O, and C atoms are labeled cyan, red, and yellow,respectively. (C) In situ XAS and XES spectra before (red) and after exposureto 10,000 ppm of CO2 (yellow) in 50% rh at 250 °C. To underline the changesthe area under the experimental curves was colored. We observe a whitelinesharpening, a decrease of the feature at −10 eV, and at the same time anincrease at the feature at −50 eV. This points at the oxidative character ofCO2 and additional oxygen in the La neighborhood. (Inset) Schematic ofpossible carbonate species forming at the surface of La2O2CO3.

15806 | www.pnas.org/cgi/doi/10.1073/pnas.1516192113 Hirsch et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

6, 2

020

respectively. Even though the sensor’s resistance upon exposurefrom dry air to 50% rh decreases 50-fold, the XAS spectra ofthe sensor measured in dry and 50% rh air are identical as shownin Fig. S7A. Thus, we conclude that the La is not an adsorptionsite for water. Upon exposure to 10,000 ppm of CO2 in 50% rh,we notice in the XAS spectra an increase of the intensity andsharpening of the whiteline, whereas other pre- and postedgefeatures do not change, as shown in Fig. S7B. These, by analogy toprevious HERFD XAS studies on Ce (62), point to the oxidativecharacter of CO2 adsorption on La2O2CO3, namely CO2 moleculesare acting as electron acceptors. We ascribe the changes of thewhiteline to the presence of additional oxygen in the direct vicinityof La. The in situ vtc XES measurements further confirm thatCO2 interacts directly with La sites and the electronic structureis rearranged because of an additional ligand, Fig. S7C; We ob-serve an increase of the states at −40 eV and a decrease of thefeature at −10 eV, which we assume to consist of La d- and O p-states. We conclude that CO2 adsorbed at La2O2CO3 as surfacecarbonates, as schematically depicted in Fig. 4. Remarkably, theXAS studies reveal an oxidative character, whereas the resistancemeasurements point at reducing character of CO2 adsorption atLa2O2CO3 in humid conditions. We note that XAS selectivelyprobes the adsorption at La site, whereas the resistance of thesensor is not selective to particular reaction or site, but insteadprobes the net charge transfer between molecules (O2, CO2, H2O)and surface of a solid (57). This further underlines the advantageof in situ XAS/XES studies for investigating the reactions mech-anism in real conditions.

ConclusionsHere, we present complementary X-ray diffraction, absorption,and emission studies of lanthanum oxycarbonate nanoparticles.To calculate the accurate XAS spectra with FEFF code a precisedetermination of atomic positions is required. Using the PXRDpattern we determine the atomic structure of monoclinic type IaLa2O2CO3. Eventually, this information is the foundation forthe FEFF calculations, which reproduce the measured XASspectrum very well. Moreover, we experimentally verify that theLa d-states are partially occupied and the empty fraction formsthe conduction band, the occupied fraction forms together withO p-states the valence bands, and the electronic d-DOS bandgap between them is 3.7 eV. To further elucidate the role of theoxycarbonate ligands an advanced density functional theorycalculation of XES spectrum is needed (12). Herein the com-bination of HERFD XAS and vtc XES techniques allows us toin situ visualize the charge transfer between relatively inertcarbon dioxide and La2O2CO3. It reveals changes of both oc-cupied (La d and O p) and unoccupied states (La d) upon the

interaction with CO2; this information is not accessible withother techniques. The CO2 sensing mechanism in humid air ishighly complex; nevertheless, our in situ results show that La is theadsorption site for carbon dioxide but not for water. In the future,the soft-range XAS/XES studies at O and C edges could possiblyanswer the question concerning sites for water and oxygen ad-sorption (63, 64), however such an experiment would be verychallenging because both elements are present in La2O2CO3 aswell as in the ambient atmosphere.

Experimental MethodsSynthesis. All reagents were used without further purification. La(OH)3 nano-particles were synthesized in a microwave reactor through heating lanthanumisopropoxide in acetophenone at 200 °C for 20 min. Afterward the sampleswere washed and dried in air. La2O2CO3 was prepared by heating this sample at500 °C for 2 h under air.

Structure Determination. The PXRD of type Ia La2O2CO3 was indexed usingTREOR (26) implemented in CMPR (27). The space group was adjusted (35)and the structure solved with EXPO2013 software (28). The lattice parame-ters were refined with GSAS (29–34).

X-Ray Spectroscopy. HERFD XAS and vtc XES experiments were carried out atID26 at the European Synchrotron Radiation Facility (ESRF), which is equippedwith an X-ray emission spectrometer (65, 66). For the HERFD XAS experimentsthe incident energy was scanned over the La L3 edge and the La Lα1 fluo-rescence line was recorded. In the vtc XES experiments the incident energywas fixed to the maximum of the whiteline energy and the spectrometerrecorded emission just below this incident energy. Powders were pressed inpellets or directly deposited on the alumina substrate, which was equippedwith Pt electrodes. The sensors were measured under working conditions inhumid air at 250 °C with pulses of 10,000 ppm CO2 in an in situ cell (16).

FEFF Calculations. The spectra were calculated with the FEFF9 program package(67, 68). The calculations for the self-consisting potential were performed in-cluding atoms within a sphere of 4 Å. For the full multiple scattering calcula-tions, atoms within a sphere of 4 Å and 4.5 Å for La(OH)3 and for La2O2CO3,respectively, were included.

Characterization Techniques. Phase composition and phase purity were in-vestigated with PXRD on a Panalytical diffractometer and FTIR on a Brukerattenuated total reflection infrared (ATR-IR) spectrometer. HRTEM micro-graphs were recorded with a Phillips Tecnai F30 electron microscope.

Detailed information about the synthesis, gas-sensing conditions, the X-rayspectroscopy measurements, the FEFF calculations, and sample characteriza-tion techniques are described in SI Experimental Methods.

ACKNOWLEDGMENTS. We thank the Eidgenössische Technische HochschuleZürich (ETH 2813-1) and the Swiss National Science Foundation(200021_137637 and 200021_1400581) for financial support and the Euro-pean Synchrotron Radiation Facility for the beamtime allocation.

1. Furukawa H, et al. (2010) Ultrahigh porosity in metal-organic frameworks. Science

329(5990):424–428.2. Millward AR, Yaghi OM (2005) Metal-organic frameworks with exceptionally high

capacity for storage of carbon dioxide at room temperature. J Am Chem Soc 127(51):

17998–17999.3. Goldwasser MR, et al. (2003) Perovskites as catalysts precursors: CO2 reforming of CH4

on Ln1−xCaxRu0.8Ni0.2O3 (Ln = La, Sm, Nd). Appl Catal, A 255(1):45–57.4. Xue D-X, et al. (2013) Tunable rare-earth fcu-MOFs: A platform for systematic enhance-

ment of CO2 adsorption energetics and uptake. J Am Chem Soc 135(20):7660–7667.5. Alezi D, et al. (2015) Quest for highly connected metal-organic framework platforms:

Rare-earth polynuclear clusters versatility meets net topology needs. J Am Chem Soc

137(16):5421–5430.6. Bhavani AG, Kim WY, Lee JS (2013) Barium substituted lanthanum manganite pe-

rovskite for CO2 reforming of methane. ACS Catal 3(7):1537–1544.7. Machida M, Yabunaka J-i, Kijima T (2000) Synthesis and photocatalytic property of

layered perovskite tantalates, RbLnTa2O7 (Ln = La, Pr, Nd, and Sm). ChemMater 12(3):

812–817.8. Nyman M, et al. (2009) Unique LaTaO4 polymorph for multiple energy applications.

Chem Mater 21(19):4731–4737.9. Alonso-Mori R, et al. (2012) Energy-dispersive X-ray emission spectroscopy using an

X-ray free-electron laser in a shot-by-shot mode. Proc Natl Acad Sci USA 109(47):

19103–19107.

10. Schmitt T, de Groot FMF, Rubensson J-E (2014) Prospects of high-resolution resonantX-ray inelastic scattering studies on solid materials, liquids and gases at diffraction-limited storage rings. J Synchrotron Radiat 21(Pt 5):1065–1076.

11. Glatzel P, Singh J, Kvashnina KO, van Bokhoven JA (2010) In situ characterization ofthe 5d density of states of Pt nanoparticles upon adsorption of CO. J Am Chem Soc132(8):2555–2557.

12. Smolentsev N, Sikora M, Soldatov AV, Kvashnina KO, Glatzel P (2011) Spin-orbitsensitive hard x-ray probe of the occupied and unoccupied 5d density of states. PhysRev B 84(23):235113.

13. Singh J, Lamberti C, van Bokhoven JA (2010) Advanced X-ray absorption and emissionspectroscopy: In situ catalytic studies. Chem Soc Rev 39(12):4754–4766.

14. Merte LR, et al. (2012) Electrochemical oxidation of size-selected Pt nanoparticlesstudied using in situ high-energy-resolution X-ray absorption spectroscopy. ACS Catal2(11):2371–2376.

15. Safonova OV, et al. (2005) Characterization of the H2 sensing mechanism of Pd-pro-moted SnO2 by XAS in operando conditions. Chem Commun (41):5202–5204.

16. Koziej D, et al. (2009) Operando X-ray absorption spectroscopy studies on Pd-SnO2

based sensors. Phys Chem Chem Phys 11(38):8620–8625.17. Boubnov A, et al. (2014) Selective catalytic reduction of NO over Fe-ZSM-5: Mecha-

nistic insights by operando HERFD-XANES and valence-to-core X-ray emission spec-troscopy. J Am Chem Soc 136(37):13006–13015.

18. Radu D, et al. (2008) Geometric and electronic structure of α-oxygen sites in Mn-ZSM-5 zeolites. J Phys Chem C 112(32):12409–12416.

Hirsch et al. PNAS | December 29, 2015 | vol. 112 | no. 52 | 15807

APP

LIED

PHYS

ICAL

SCIENCE

S

Dow

nloa

ded

by g

uest

on

Mar

ch 2

6, 2

020

19. Djerdj I, et al. (2009) Neodymium dioxide carbonate as a sensing layer for chemo-resistive CO2 sensing. Chem Mater 21(22):5375–5381.

20. Willa C, Yuan J, Niederberger M, Koziej D (2015) When nanoparticles meet poly(ionicliquid)s: Chemoresistive CO2 sensing at room temperature. Adv Funct Mater 25(17):2537–2542.

21. Haensch A, Koziej D, Niederberger M, Barsan N, Weimar U (2010) Rare earth oxy-carbonates as a material class for chemoresistive CO2 gas sensors. Procedia Eng 5(0):139–142.

22. Schneider J, Jia H, Muckerman JT, Fujita E (2012) Thermodynamics and kinetics of CO2,CO, and H+ binding to the metal centre of CO2 reduction catalysts. Chem Soc Rev41(6):2036–2051.

23. Turcotte RP, Sawyer JO, Eyring L (1969) Rare earth dioxymonocarbonates and theirdecomposition. Inorg Chem 8(2):238–246.

24. Olafsen A, Fjellvag H (1999) Synthesis of rare earth oxide carbonates and thermalstability of Nd2O2CO3 II. J Mater Chem 9(10):2697–2702.

25. Christensen AN (1970) Hydrothermal preparation of neodymium oxide carbonate.The location of the carbonate ion in the structure of Nd2O2CO3. Acta Chem Scand 24:2440–2446.

26. Werner P-E, Eriksson L, Westdahl M (1985) TREOR, a semi-exhaustive trial-and-errorpowder indexing program for all symmetries. J Appl Cryst 18(5):367–370.

27. Toby B (2005) CMPR - a powder diffraction toolkit. J Appl Cryst 38(6):1040–1041.28. Altomare A, et al. (2013) EXPO2013: A kit of tools for phasing crystal structures from

powder data. J Appl Cryst 46(4):1231–1235.29. Rietveld H (1969) A profile refinement method for nuclear and magnetic structures.

J Appl Cryst 2(2):65–71.30. Larson AC, Von Dreele RB (1994) General Structure Analysis System (GSAS). (Los

Alamos Natl Lab, Los Alamos, NM), Report LAUR 86-748.31. Toby B (2001) EXPGUI, a graphical user interface for GSAS. J Appl Cryst 34(2):210–213.32. Young RA (1993) The Rietveld Method (Oxford Univ Press, Oxford).33. Suortti P (1972) Effects of porosity and surface roughness on the X-ray intensity re-

flected from a powder specimen. J Appl Cryst 5(5):325–331.34. Von Dreele R (1997) Quantitative texture analysis by Rietveld refinement. J Appl Cryst

30(4):517–525.35. cctbx - Change space group setting. cci.lbl.gov/cctbx/change_setting.html. Accessed

December 9, 2014.36. Li H, et al. (2013) The synthesis of glycerol carbonate from glycerol and CO2 over

La2O2CO3-ZnO catalysts. Catal Sci Technol 3(10):2801–2809.37. Cornaglia LM, Múnera J, Irusta S, Lombardo EA (2004) Raman studies of Rh and Pt on

La2O3 catalysts used in a membrane reactor for hydrogen production. Appl Catal, A263(1):91–101.

38. Múnera JF, et al. (2007) Kinetics and reaction pathway of the CO2 reforming ofmethane on Rh supported on lanthanum-based solid. J Catal 245(1):25–34.

39. Krause MO, Oliver JH (1979) Natural widths of atomic K and L levels, Kα X‐ray linesand several KLL Auger lines. J Phys Chem Ref Data 8(2):329–338.

40. Journel L, et al. (2002) Resonant inelastic x-ray scattering at the lanthanum L3 edge.Phys Rev B 66(4):045106.

41. Aritani H, Yamada H, Yamamoto T, Tanaka T, Imamura S (2001) XANES study of Li-MgO and Li-La2O3-MgO catalysts for oxidative coupling of methane. J SynchrotronRadiat 8(Pt 2):593–595.

42. Asakura H, Shishido T, Teramura K, Tanaka T (2014) Local structure and La L1 and L3-edge XANES spectra of lanthanum complex oxides. Inorg Chem 53(12):6048–6053.

43. Peacock PW, Robertson J (2002) Band offsets and Schottky barrier heights of highdielectric constant oxides. J Appl Phys 92(8):4712–4721.

44. Singh N, Saini SM, Nautiyal T, Auluck S (2006) Electronic structure and optical prop-erties of rare earth sesquioxides (R2O3, R=La, Pr, and Nd). J Appl Phys 100(8):083525.

45. Kvashnina KO, Butorin SM, Glatzel P (2011) Direct study of the f-electron configu-ration in lanthanide systems. J Anal At Spectrom 26(6):1265–1272.

46. Glatzel P, et al. (2005) Range-extended EXAFS at the L edge of rare earths using high-energy-resolution fluorescence detection: A study of La in LaOCl. Phys Rev B 72(1):014117.

47. de Groot F, Vankó G, Glatzel P (2009) The 1s x-ray absorption pre-edge structures intransition metal oxides. J Phys Condens Matter 21(10):104207.

48. Sunding MF, et al. (2011) XPS characterisation of in situ treated lanthanum oxide andhydroxide using tailored charge referencing and peak fitting procedures. J ElectronSpectrosc Relat Phenom 184(7):399–409.

49. Mikami M, Nakamura S (2006) Electronic structure of rare-earth sesquioxides andoxysulfides. J Alloys Compd 408–412(0):687–692.

50. Jiang H, Gomez-Abal RI, Rinke P, Scheffler M (2009) Localized and itinerant states inlanthanide oxides united by GW @ LDA+U. Phys Rev Lett 102(12):126403.

51. Szlachetko J, Pichler M, Pergolesi D, Sa J, Lippert T (2014) Determination of conduc-tion and valence band electronic structure of La2Ti2O7 thin film. RSC Advances 4(22):11420–11422.

52. Szlachetko J, Sa J (2013) Rational design of oxynitride materials: From theory to ex-periment. CrystEngComm 15(14):2583–2587.

53. Chiou JW, et al. (2008) Mg-induced increase of band gap in Zn1−xMgxO nanorodsrevealed by X-ray absorption and emission spectroscopy. J Appl Phys 104(1):013709.

54. Bär M, et al. (2008) Depth-resolved band gap in Cu(In,Ga)(S,Se)2 thin films. Appl PhysLett 93(24):244103.

55. Mu Q, Wang Y (2011) Synthesis, characterization, shape-preserved transformation,and optical properties of La(OH)3, La2O2CO3, and La2O3 nanorods. J Alloys Compd509(2):396–401.

56. Mu Q, Wang Y (2011) A simple method to prepare Ln(OH)3 (Ln = La, Sm, Tb, Eu, andGd) nanorods using CTAB micelle solution and their room temperature photo-luminescence properties. J Alloys Compd 509(5):2060–2065.

57. Barsan N, Simion C, Heine T, Pokhrel S, Weimar U (2010) Modeling of sensing andtransduction for p-type semiconducting metal oxide based gas sensors. J Electroceram25(1):11–19.

58. Erbil A, Cargill III, Frahm R, Boehme RF (1988) Total-electron-yield current measure-ments for near-surface extended x-ray-absorption fine structure. Phys Rev B CondensMatter 37(5):2450–2464.

59. Vlachos D, Craven AJ, McComb DW (2005) Specimen charging in X-ray absorptionspectroscopy: Correction of total electron yield data from stabilized zirconia in theenergy range 250-915 eV. J Synchrotron Radiat 12(Pt 2):224–233.

60. Carroll KJ, et al. (2013) Probing the electrode/electrolyte interface in the lithium ex-cess layered oxide Li1.2Ni0.2Mn0.6O2. Phys Chem Chem Phys 15(26):11128–11138.

61. Schroeder S, Moggridge G, Lambert R, Rayment T (1998) Spectroscopy for SurfaceScience (Wiley, Hoboken, NJ).

62. Safonova OV, et al. (2014) Electronic and geometric structure of Ce3+ forming underreducing conditions in shaped ceria nanoparticles promoted by platinum. J PhysChem C 118(4):1974–1982.

63. Albrecht PM, Jiang D-e, Mullins DR (2014) CO2 Adsorption as a flat-lying, tridentatecarbonate on CeO2(100). J Phys Chem C 118(17):9042–9050.

64. Dell’Angela M, et al. (2013) Real-time observation of surface bond breaking with anx-ray laser. Science 339(6125):1302–1305.

65. Glatzel P, Bergmann U (2005) High resolution 1s core hole X-ray spectroscopy in 3dtransition metal complexes—electronic and structural information. Coord Chem Rev249(1–2):65–95.

66. Glatzel P, Sikora M, Smolentsev G, Fernández-García M (2009) Hard X-ray photon-inphoton-out spectroscopy. Catal Today 145(3–4):294–299.

67. Rehr J, Albers R (2000) Theoretical approaches to x-ray absorption fine structure. RevMod Phys 72(3):621–654.

68. Rehr JJ, Kas JJ, Vila FD, Prange MP, Jorissen K (2010) Parameter-free calculations ofX-ray spectra with FEFF9. Phys Chem Chem Phys 12(21):5503–5513.

15808 | www.pnas.org/cgi/doi/10.1073/pnas.1516192113 Hirsch et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

6, 2

020