Macroscopic and Molecular Insights from CO Adsorption on NaY

May 7, 2012 - Laboratoire Interdisciplinaire Carnot de Bourgogne, UMR 5209 CNRS ... Anthony Ballandras , Guy Weber , Christian Paulin , Jean-Pierre Be...
0 downloads 0 Views 399KB Size
Subscriber access provided by UNIVERSITY OF SASKATCHEWAN LIBRARY

Article

Macroscopic and Molecular Insights from CO Adsorption on NaY Zeolite: A Combined FTIR and Manometric Study Olivier Cairon, and Jean Pierre Bellat J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp302883p • Publication Date (Web): 07 May 2012 Downloaded from http://pubs.acs.org on May 9, 2012

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Macroscopic and Molecular Insights from CO Adsorption on NaY zeolite: A combined FTIR and Manometric Study. Olivier Cairon* and Jean-Pierre Bellat† * UFR Sciences & Techniques, Université de Pau BP 1155, 64013 Pau, France † Laboratoire Interdisciplinaire Carnot de Bourgogne, UMR 5209 CNRS, Université de Bourgogne, 9 av. Savary, BP 47870, 21078 Dijon, France [email protected], [email protected] RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required according to the journal that you are submitting your paper to)

ABSTRACT

This survey combines both quantitative and IR molecular descriptions and aims to provide new insights for the description of CO adsorption on NaY zeolite at 77 K. Quantitative measurements of the number of CO molecules trapped in the microporous super cage are compared to the corresponding IR spectra of CO as adsorbed species. We demonstrate that polycarbonyls formed during the completion of the accessible SII Na+ coordinative vacancies result in the formation of mono-, di- and tri-carbonyls but not consecutively. Quantitative analysis and measurements of the CO molecules that are adsorbed prove that polycarbonyls coexist with different proportions over the adsorption step in line with previous ACS Paragon Plus Environment

1

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 16

theoretical prediction.1 Moreover, we establish that polycarbonyls formation settles rapidly and stops, although the completion of the SII Na+ coordinative vacancies has not been fully achieved. The formation of CO in the pseudo-liquid phase causes this adsorption limitation and distinction within adsorbed species must be made between species that are truly Na+-coordinated and those that are adsorbed in a condensed state in the confined micropores. As a result, this combined analysis reveals that the sole manometric measurements resulting in the macroscopic CO isotherm overestimate the number of ‘true’ coordinated species.

KEYWORDS Adsorption – Isotherm – NaY – Cationic Zeolites - IR Spectroscopy – CO – Probe Molecules

INTRODUCTION Cationic alkali-exchanged zeolites are synthetic microporous materials widely used as selective adsorbents, ion exchangers, and heterogeneous catalysts. Their use in the separation processes mainly concerns gas/solid adsorption.2 Indeed, zeolites’ microporous structure combined with the affinity for guest molecules that contain cations have made these materials a very efficient molecular sieve for gas mixtures. Therefore, gas separation performances of alkali-exchanged zeolites strongly depend on both their adsorption affinity and their capacity for a given gas. To assess zeolites’ performances in this field, experimental adsorption isotherms are usually performed by gravimetric or manometric experiments.2,3 In spite of the very informative insights deduced from these experimental isotherms, such an approach suffers from an understanding limited to the macroscopic-scale because only a global evaluation of the adsorbed amount of gas is reached. Addressing the molecular understanding of the adsorption process is of crucial importance for improving knowledge of the role of alkali cations relating to gas adsorption capacity and affinity. Moreover, this approach may offer breakthroughs for tailoring more powerful adsorbents and more selective porous materials.

ACS Paragon Plus Environment

2

Page 3 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Among physico-chemical techniques, FTIR spectroscopy of small-adsorbed probe molecules has proven to offer many advantages over the others, especially for molecular investigations of gas adsorption on cationic zeolites.4,5 Nevertheless, correlating the IR spectra of adsorbed species and quantitative measurements of gas adsorption remains a complex issue, especially where more than one adsorbed species is detected. Therefore, the combination of the classical manometric experiments that provide quantitative measurements and IR investigations that enable molecular insights on adsorbed species should enlarge and refine our understanding of gas adsorption on both macroscopic and molecular scales. However, such combined investigations first require that appropriate experiments be conducted in fully comparable conditions, whatever the size scale may be. In this paper, we focus on the analysis of CO adsorption on NaY Faujasite investigated by means of both FTIR and volumetric measurements. We aim to examine how these two approaches, when combined, can lead to new qualitative as well as quantitative insights on gas adsorption for porous materials.

EXPERIMENTAL SECTION IR measurements: Progressive CO adsorption at 77K was carried out for a self-supporting thin pellet of fully sodium-exchanged Na56Y zeolite (Union Carbide, Si/Al ~ 2.5, anhydrous weight m = 3.61 mg). Both manometric and molecular descriptions were simultaneously obtained in the IR cell. IR spectra of NaY zeolite were recorded with a resolution of 2 cm-1. The pellet of the NaY zeolites was first activated with a heating rate of 2 K min-1 up to 650 K and was held for 2 hours in a home-made designed cell connected to a vacuum adsorption system (P < 10-3 Pa). This activation totally dehydrated the Na56Y zeolite (no residual water was IR detected). After activation, the wafer was cooled down progressively to liquid nitrogen temperature (77 K) and the background spectrum of the NaY zeolite was measured. The CO molecules were introduced stepwise from a control volume (2 cm3) at 77 K by doses ranging from 0.4 to 40 mmol of CO. For each added dose, both the IR spectrum and the corresponding equilibrium pressure (PCO) were recorded thanks to a capacitive gauge (10-3 mbar of exactness). Molecular ACS Paragon Plus Environment

3

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 16

investigations consisted of analyzing IR spectra through IR signatures of polycarbonyls (as recently evidenced). Isotherm construction in the IR cell was derived from macroscopic PCO measurements. Calculations were made using adsorbed CO quantities that we deduced by correlating each equilibrium pressure (PCO) to the CO doses added. For this task, the cell pressure was previously calibrated under the same conditions, but without a pellet. The number of adsorbed CO molecules were calculated and expressed per supercage based on the following data: molecular weight of anhydrous Na56Y = 12752 g/mol, 8 super cages per unit cell, a pellet having a surface area of S= 1.327 cm2 and anhydrous weight of m = 3.61 mg. As a result, both molecular and macroscopic investigations were made in one pot in the IR cell and joined for a quantitative assessment with molecular accuracy. Macroscopic classical isotherm: Note that the classical adsorption isotherm was obtained using an independent and conventional manometric montage while the IR cell was equipped with capacitive gauges allowing both manometric and IR measurements. An conventional experimental montage was used with a similar protocol for IR manometric measurements (activation and CO admission).6 However, due to their respective experimental approaches and devices, the following differences must be underlined: a greater amount of zeolite was used (m = 263 mg), a smaller cell was used (volume = 16.06 cm3) while the controlled volume for CO admission was larger (180.34 cm3). Capacitive gauge were of 10-4 mbar of exactness.

RESULTS AND DISCUSSION As noted above, CO adsorption on NaY was chosen as to assess the feasibility and relevance of such combined investigation on cationic zeolites. We adopted the following strategy. First, macroscopic isotherms of CO adsorption on NaY were obtained by classical manometric experiments and compared to those deduced from manometric measurements conducted in the FTIR cell. As a result, the total number of CO adsorbed molecules was calculated and expressed per each supercage (s.c.) that contained the four SII Na+ cations accessible solely to CO molecules.7 Accordingly, we reported results from IR and classical manometric methods are reported, Figure 1. ACS Paragon Plus Environment

4

Page 5 of 16

16 N CO (per s.c.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

14 12 10

C

8

B

6 A

4 2

PCO (Pa) 0

0

20

40

60

80

100

Figure 1: Number of adsorbed CO molecules per supercage (s.c) deduced from manometric measurements: with a classical device (x), made in the IR cell (○). The sets A, B and C correspond to 4, 8 and 11.4 CO molecules per supercage, respectively. While the classical manometric experiment accounts for about 14 CO molecules per supercage (s.c.) to fully filling supercage, manometric measurements in the IR cell shows up to 11.7 CO. We observed these values at a pressure of 15 Pa, when the plateau is formed (Fig. 1). Such a difference was repeatedly observed independent of the weight of the NaY pellet used in the IR cell but also for classical measurements conducted with bare or pressed powders of NaY. As a first conclusion, quantitative measurements of CO quantities deduced from the manometric classical method always exceed those obtained in the IR cell. It seems that this discrepancy is more likely due to the use of different experimental devices and montages and is repeatedly obtained (see experimental part above). In the second part of our strategy, we recorded IR spectra of adsorbed species for each of the CO doses as correlated to the manometric isothermal and quantitative adsorption of CO conducted in the IR cell (Fig. 1). In this way the impact of each CO dose was assessed with both molecular and quantitative

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 16

accuracies. To better understand the results developed hereafter, let us briefly consider the main molecular insights that account for CO adsorption on NaY. The well-known structure of NaY Faujasite is composed of silica and alumina tetrahedra joined together to form the so-called sodalite units also termed β-cages. The connection of these sodalite units is ensured by hexagonal prisms thus displaying larger porous cages, also termed supercages (s.c.). The presence of Al atoms introduces negative charges to the structure that are balanced by Na+ extraframework cations spreading in privileged stable positions in the supercages but also in β cages or in hexagonal prisms. The distribution of sodium cations on the above-mentioned cages is well established in the dehydrated NaY Faujasite.7 As a result, only four Na+ extra-framework cations populate the SII positions of each supercage. Moreover, due to steric hindrance, CO adsorption only occurs in the supercage since CO can penetrate neither β cages nor hexagonal prisms. By means of IR spectroscopy, we recently augmented and refined earlier worksdevoted to CO adsorption on NaY.8-10 Thus, we evidenced new IR signatures during CO adsorption and demonstrated that progressive CO adsorption results from the completion of the coordination vacancies of the SII Na+ cations (Scheme 1). Accordingly, mono, di- and tri-carbonyls could be formed as carbonyl adsorbed species -Na+(CO), Na+(CO)2 and Na+(CO)3. These carbonyls were IR evidenced through their respective vibrating average νCO positions ca 2171, 2163 and 2150 cm-1, respectively. 8-10

supercage CO CO

CO Na

+

O atoms Al or Si atoms β - cage

ACS Paragon Plus Environment

6

Page 7 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Scheme 1: Schematic representation of SII Na+ cation position in NaY Faujasite interacting with the CO probe. Mono-, di- and tri-carbonyls as possible adsorbed species.

Moreover, we also showed that depending on the CO pressure reached in the cell progressive CO adsorption takes place according to a set of four equilibrium reactions. Based on the analysis of the appearance of each of the three adsorbed carbonyl species, these four equilibrium reactions have recently been proposed: 9,10 Na+ + CO ↔ Na+(CO)

(1)

Na+ + 2CO ↔ Na+(CO)2

(2)

Na+(CO) + CO ↔ Na+(CO)2

(3)

Na+ + 3CO ↔ Na+(CO)3

(4)

Note that tri-carbonyls’ formation coming from di-carbonyls’ conversion was not experimentally observed as possible species resulting from the addition of one CO molecule to existing dicarbonyls.9,10. Nevertheless, it was not possible to compare these new IR insights with further quantitative considerations because the CO massif is complex and embraces the formation of different polycarbonyls species where multiple composite bands appear. As a result, quantitative assessment of the contribution of each carbonyl as adsorbed species to the total amount of adsorbed CO is not obviously accessible by IR spectra analysis. Such an assessment would require first decomposing the massif according each band contribution and then determine their respective molar adsorption coefficient.11,12 Although it would make matters easier, no quantitative data of CO adsorption on NaY deduced from manometric or gravimetric experiments have been reported in the literature. In the second part of our strategy, we attempted to overcome these quantitative issues by combining both manometric and IR measurements in the IR cell. IR spectra recorded for progressive CO additions (from 0.4 to 40 mmol.) are reported in Figure 2.

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry

2162

3

2 Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 16

2171 0.2

1

2180

2160 Wavenumbers (cm-1)

2140

2120

Figure 2: FTIR spectra of progressive CO adsorption on NaY at 77 K. The three main steps 1, 2 and 3 are for low, medium and high CO loadings, respectively.

Three striking steps for CO capture are discernable, depending on the amount of the CO probe that is admitted (all three steps are connected to the four above-mentioned equilibrium reactions). The first step accounted for the Na+(CO) mono-carbonyl formation of reaction (1). Second step is obtained when increasing CO pressure (i.e. CO doses) and resulted in di-carbonyls formation from both direct di-carbonyls formation (reaction 2) and mono-carbonyls conversion (reaction 3). Finally, the third step clearly accounted for the predominant di-carbonyls formation (although the band’s width increased close to the 2150 cm-1 position which could reflect tri-carbonyls formation as previously suggested).8-10 A more accurate analysis of the adsorbed species formed during this last step was strongly hindered by the presence of CO in a pseudo liquid phase. This was detectable by the growing large band around the 2140 cm-1 position and caused a poor resolution of the massif. Let us now combine this qualitative molecular description with the macroscopic and quantitative manometric results as depicted above in Fig. 1. For this task and for sake of brevity, we have highlighted some remarkable quantities of adsorbed CO molecules and connected them to their related IR spectra (displayed below in Fig. 3). Finally, we have focused on the deducible quantitative analysis. These

ACS Paragon Plus Environment

8

Page 9 of 16

striking sets correspond to the adsorption of 4, 8, and 11.4 CO molecules per supercage and are termed A, B and C, respectively (as reported in Fig. 1). Note that each supercage contains 4 Na+ cations, all located in SII positions (the only positions accessible to the CO probe).

(10.9-11.4) C

Abs orb anc e

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(7.9-8.5) B 0.2

(3.9-4.2) A

2180

2160

2140

2120

Wavenumbers (cm-1)

Figure 3: FTIR spectra corresponding to the striking adsorbed CO doses (sets A, B, C) as reported in Fig. 1. Exact CO adsorbed amounts per supercage are indicated in brackets. At the end of the A set, IR spectra were related to the adsorption of 4 CO molecules per supercage according to the quantitative measurements reported in Fig. 1. As noted above, 4 Na+ cations populated the SII positions in the supercage. Assuming that CO adsorption occurs through a progressive and stepwise process as consensually accepted,13,14 step A accounts for one CO molecule being captured by each of the four SII Na+ cations. According to this assumption, the IR band (2171 cm-1) recorded in the A set would represent the achievement of mono-carbonyls formation for each supercage and thus this band should have reached its maximum intensity. Further CO additions contradicted this assumption and made the step by step process of CO capture strongly unlikely. Indeed, when last IR spectra of the A set was taken as background, spectral subtractions (made for further CO doses, A to B run, and reported in Fig. 4) revealed that for the two first subtractions a positive band can be observed close to the monocarbonyls position (2171 cm-1). Thus, mono-carbonyls formation continued beyond the A set. This

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry

definitively confirmed that during the A set, CO capture does not result from mono-carbonyls alone, but also from other adsorbed species that are formed concomitantly. On the other hand, the simultaneous emergence of the massif around the 2163 cm-1 position (dicarbonyls) and the negative part around 2171 cm-1 revealed that the band of mono-carbonyls disappeared progressively in relation to di-carbonyls formation, Fig. 4. Thus, it is clearly demonstrated that during the first step A the 4 CO molecules trapped in each supercage containing 4 Na+ cations did not result from a homogeneous and concerted process uniquely involving mono-carbonyls formation. Accordingly, to successfully match the 4 CO molecules adsorbed during step A as indicated by the manometric measurements, other adsorbed species -likely di-carbonyls- were formed concomitantly. These quantitative findings conform to our previous demonstration that di-carbonyls can be formed even when mono-carbonyls’ formation is not achieved.8,9 2163

Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 16

0.2 2171

2200

2180

2160

2140

2120

Wavenumbers (cm-1)

Figure 4: FTIR difference spectra for the CO additions during the A to B run. The IR spectrum corresponding to four CO per supercage is taken as background spectra. Moreover, these spectral insights of the A to B run where the concomitant mono- and di-carbonyls formations occur potentially explains the absence of any clear isosbestic point. If the mono to dicarbonyls conversion had occur only (reaction 3), an isosbestic pointwould have been present. Instead of

ACS Paragon Plus Environment

10

Page 11 of 16

a symmetric band, a trailing band separate from the mono and di-carbonyls IR positions is emerged, Fig. 4. Using the same approach, accurate analysis of the second step B at around 8 CO molecules per supercage completes the previous description of the A to B run. According to quantitative measurements (Fig. 1) 8 CO were trapped in the s.c. in the B set. Once again, a stepwise CO capture should result in the complete formation of di-carbonyls in the 4 SII Na+ cations. To assess this hypothesis, the approach developed above was subsequently applied to the B to C run. Spectral subtractions taking IR spectra of the B set as background are depicted in Fig. 5 for further CO doses.

Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

2200

0.2

2180

2160

2140

2120

Wavenumbers (cm-1)

Figure 5: FTIR difference spectra for the CO additions during the B to C run. The IR spectrum corresponding to eight CO per supercage is taken as background spectrum. The negative band at 2171 cm-1 clearly accounts for the conversion of mono-carbonyls into dicarbonyls. Concomitantly, a positive band first emerged close to the 2163 cm-1 position and was for the last CO doses red-shifted to a position lower than 2160 cm-1. We believe these features account for dicarbonyls formation (positive band at 2163 cm-1) that originated from reaction 3 of mono- to diconversion (negative band at 2171 cm-1). According to these results, both mono- and di-carbonyls were present in the B set. As exposed in Fig. 5, di-carbonyls formation was far from being achieved in the B set since it continued during the B to C run. From a quantitative point of view, since both mono- and dicarbonyls were present in the set B, it implies that other adsorbed species were also formed to ACS Paragon Plus Environment

11

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 16

successfully match the eight adsorbed CO molecules as indicated by the manometric measurements (Fig. 1). To clarify this, let us reexamine the spectral subtractions reported in Fig. 4. For the last spectra reported in Fig. 4, a large tail was observed in the lowest wavenumber part of the IR massif. Although no clear maximum is visible the tail position at around 2150 cm-1 strongly suggests the presence of tricarbonyls as reported previously.8,9 The above quantitative argument fully supports the presence of tricarbonyls in the set B. Let us now examine the step C (11.4 CO per supercage). Quantitatively, the step C clearly accounts for the beginning of CO saturation (see Fig. 1). In Fig. 5, spectral features clearly show that after the two first spectra (similar to those reported in Fig. 4) the next spectra develop new trends. First, the trailing shape that was barely visible around the 2150 cm-1 position (Fig. 4) has increased and is well highlighted in Fig. 5. Second, this trailing band is clearly accompanied by the emergence of a new band at around 2140 cm-1. This recognizable band is assigned to CO in a pseudo liquid phase. As a result, the overall position of the main band was progressively red-shifted and can now be detected at 2159 cm-1, a position that must be compared to the 2163 cm-1 band due to the unique presence of di-carbonyls. According to the arguments developed above, the presence of tri-carbonyls would strongly corroborate both this red shift (and the resulting band at 2159 cm-1) as well as the above-mentioned quantitative considerations stating that at the step C, 11.4 CO molecules per s.c. are adsorbed. Indeed, these 11.4 CO can not result in the near completion of the 4 SII Na+ cations having mainly tri-carbonyls as adsorbed species. Such an assumption implies that the resulting IR massif in Fig. 5 would have been very similar to a the IR signature of a unique and symmetric band closely located near 2150 cm-1 as reported for tricarbonyls.8-10 The above-mentioned spectral features invalidate such an assumption regarding tricarbonyls as unique adsorbed species. Moreover, closer inspection of the last spectra attained in the set C (Fig. 5) indicates the stagnation of di-carbonyls formation. The only band that shows a significant growth is attributed to CO in a pseudo liquid phase at around 2140 cm-1. Based on these arguments, reaching the 11.4 CO per s.c. adsorbed in the set C necessarily implies the presence of CO molecules in a pseudo liquid phase. ACS Paragon Plus Environment

12

Page 13 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

As a consequence, two classes of adsorbed species were apparent in the set C. The first was made up of adsorbed species (mono-, di-, tri-carbonyls) involving the unsaturated Na+ cations while the second was comprised of CO in a condensed phase. This phase differed from those of CO adsorbed species resulting from the completion of the SII Na+ cations vacancies. These findings thus indicate that quantitative measurements made at a macroscopic level using the manometric method do not differenciate CO molecules that are adsorbed through coordination (i.e. polycarbonyls via SII Na+ cations) from those adsorbed in a condensed state (i.e. CO in pseudo liquid phase). Consequently, the correlation of the quantitative manometric measurements with the number of SII Na+ cations that NaY contains is far from being obvious. Finally, to complete this description, the role of CO in pseudo liquid phase could strongly influence the mechanisms of CO capture via the coordination process. On one hand, CO adsorption was controlled by the four thermodynamic equilibrium involving a priori the three coordinatively adsorbed species (i.e. mono-, di- and tri-carbonyls) and depending on the pressure of CO that was reached. On the other hand, CO in pseudo liquid phase prevented any additional formation of coordinatively adsorbed species. This was quantitatively witnessed in set C just after adsorption settled to around 11.7 CO molecules per s.c. (the plateau was reached). The steric hindrance potentially induced by CO in liquid phase provides one explanation for this finding, resulting in the disruption of CO adsorption regarding the completion of vacant coordination of Na+ cations and the cessation of polycarbonyls formation. These new findings are summarized below in scheme 2 based on our analyses exposed for the sets A, B and C.

Mono PCO

Di Tri

Liq.

A B

C

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 16

Scheme 2: Schematic CO adsorption according the sets A, B and C (4, 8 and 11.4 CO per s.c.) Corresponding qualitative and semi-quantitative description of the mixture of polycarbonyls ( Mono, Di and Tri) and CO in a pseudo liquid phase as adsorbed species. CONCLUSION Description of CO adsorption on NaY zeolite obtained by a combination of quantitative manometric measurements and IR molecular investigations provides new conceptual breakthroughs regarding the technical approach as well as potential results. The two approaches, when used complementarily, make up for their respective weaknesses when used independently. For example, the complexity and laboriousness of the quantitative aspects of an IR approach take full advantage of manometric measurements. The resulting sequences of adsorbed CO probe are quantitatively assessed to finally reveal something hitherto unknown on the coexistence of different adsorbed species: the coordinative vacancies of cations does not occur in a concerted manner since species of mono-, di- or tri-carbonyls, respectively, attest to differences for the completion of Na+ cations’ vacancies. From a spectral point of view, this would explain the complexity of the observed IR massifs that potentially cover all the adsorbed species mentioned above. On the other hand, the manometric approach that remains a macroscopic assessment of quantitative aspects of gas adsorption is perfectly complemented by the IR molecular information. In this study, we showed that the total amount of gas adsorbed corresponded to two adsorption phenomena. The first, which appeared for the original doses of CO, involved the ability of cations to capture the probe. The latter phenomenon, which appeared with higher doses, concerned only the condensation of the gas in the confined space of the micropores and resulted in an overestimation of the adsorption capacity (coordinative-adsorbed quantity) of cationic zeolites that manometric measurements can not distinguish. The results layed out for this specific issue could have implications for how we understand similar systems involving other cationic porous materials with a capacity for gas adsorption or gas separation.

ACS Paragon Plus Environment

14

Page 15 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

REFERENCES (1) Hadjiivanov, K.; Ivanova, E.; Klissurski, D. Catal. Today 2001, 70, 73. (2) Yang, R.T. in Gas Separation by Adsorption Processes, Imperial College Press, London, 1997. (3) Rouquerol, F.; Rouquerol, J.; Sing, K.S.W. in Adsorption by Powders and Porous Solids, Academic Press, San Diego, 1999. (4) Knözinger, H.; Huber, S. J. Chem. Soc., Faraday Trans. 1998, 94, 2047. (5) Hadjiivanov, K.; Knözinger, H. Surface Science 2009, 603, 1629. (6) Simonot-Grange, M.H.; Bertrand, O.; Pilverdier, E.; Bellat, J.P.; C. Paulin, C. J. Thermal Anal. 1997, 48, 741. (7) Frising, T.; Leflaive, P. Micro. Meso. Mat. 2008, 114, 27 and references therein. (8) Cairon, O.; Loustaunau, A. J. Phys. Chem. C 2008, 112, 18493. (9) Cairon, O. Phys. Chem. Chem. Phys. 2010, 12, 14217. (10) Cairon, O.; Guesmi, H. Phys. Chem. Chem. Phys. 2011, 13, 11430. (11) Cairon, O.; T. Chevreau, T. J.Chem.Soc.,Faraday-Trans., 1998, 94(2), 323. (12) Cairon, O.; T. Chevreau, T.; Lavalley , J.C. J.Chem.Soc.Faraday-Trans. 1998, 94(19), 3039. (13) Bordiga, S.; Platero, E. E.; Areán, C.O.; Lamberti, C.; Zecchina, A. J. Catal. 1992, 137. (14) Hadjiivanov, K.; Knözinger, Chem. Phys. Lett., 1999, 303, 513.

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry

GRAPHIC Manometric Isotherm COper supercage

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 16

11.5 CO 8 CO

Liq. Tri Adsorbed Carbonyl Species

4 CO

Di

IR spectroscopy Mono 0

20

40

60

ACS Paragon Plus Environment

PCO (Pa) 80

16