This is the peer reviewed version of the following article:

Pérez-Carvajal J., Boix G., Imaz I., Maspoch D.. The
Imine-Based COF TpPa-1 as an Efficient Cooling Adsorbent
That Can Be Regenerated by Heat or Light. Advanced Energy
Materials, (2019). . 1901535: - . 10.1002/aenm.201901535,

which has been published in final form at
https://dx.doi.org/10.1002/aenm.201901535. This article
may be used for non-commercial purposes in accordance with
Wiley Terms and Conditions for Use of Self-Archived Versions.
The Imine-Based COF TpPa-1 as an Efficient Cooling
Adsorbent That Can Be Regenerated by Heat or Light
Javier Pérez-Carvajal, Gerard Boix, Inhar Imaz,* and Daniel Maspoch*
Javier Pérez-Carvajal, Gerard Boix, Inhar Imaz, and Daniel Maspoch:
Catalan Institute of Nanoscience and Nanotechnology (ICN2)CSIC and BIST
Campus UAB, Bellaterra08193 Barcelona, Spain
E-mail: inhar.imaz@icn2.cat; daniel.maspoch@icn2.cat
Prof. D. Maspoch:
ICREA
Pg. Lluís Companys 23, Barcelona 08010, Spain

Adsorption‐based cooling systems, which can be driven by waste heat and solar energy, are promising
alternatives to conventional, compression‐based cooling systems, as they demand less energy and emit
less CO2. The performance of adsorption‐based cooling systems relates directly to the performance of
the working pairs (sorbent–water). Accordingly, improvement of these systems relies on the continual
discovery of new sorbents that enable greater mass exchange while requiring less energy for
regeneration. Here, it is proposed that covalent‐organic frameworks (COFs) can replace traditional
sorbents for adsorption‐based cooling. In tests mimicking standard operating conditions for industry,
the imine‐based COF TpPa‐1 exhibits a regeneration temperature below 65 °C and a cooling coefficient
of performance of 0.77 – values which are comparable to those reported for the best metal–organic
framework sorbents described to date. Moreover, TpPa‐1 exhibits a photothermal effect and can be
regenerated by visible light, thereby opening the possibility for its use in solar‐driven cooling.

Increasing fuel consumption and prices, combined with anthropogenic greenhouse gas emissions, are
together driving the development of new technologies and materials to reduce society's electrical
energy demands. Already, more than 44% of all primary energy in the US residential and commercial
sectors is consumed by environmental control systems such as cooling devices, and this percentage is
predicted to increase (https://www.iea.org/weo2017/). To reverse such consumption, several
initiatives have been proposed, including use of adsorption heat transformation and storage (AHTS)
systems, such as adsorption‐cooling systems/chillers, heat pumps, and thermal batteries.1-3 These
systems employ energy‐delivery processes based on a reversible adsorption/desorption cycle of a
working fluid, whereby useful heat is released during the exothermal adsorption step and cold is
produced during the evaporation of the fluid. Among their advantages, these systems enable use of
low thermal‐energy sources (e.g., solar radiation and waste heat) for regeneration and driving energy,
and use of water as the working fluid.4 The efficiency of these processes relates directly to the
performance of the working pairs (adsorbent–adsorbate) in terms of both the level of mass exchange
and the amount of heat required for regeneration.5 Commercially6 available, thermally driven
adsorption chillers and heat pumps employ traditional porous sorbents such as silica, activated
carbon, and zeolites. However, these porous materials have limited performance, exhibiting very low
adsorption uptake in the working range (0.05–0.32% relative humidity) and requiring high
temperatures for regeneration.7-10

Metal–organic frameworks (MOFs) are an emerging class of versatile porous materials that have
recently been proposed for AHTS applications,11-14 as they can take up large amounts of water within
the abovementioned working range and show “S‐”type sorption isotherms (e.g., type IV/V). Such
isotherms are desirable for AHTS applications since their maximum working capacity falls within a very
narrow window of relative pressure (or relative humidity). Under these conditions, porous materials
can be easily regenerated upon small variations in temperature and can reversibly adsorb/desorb
water upon minor changes in relative water pressure (or relative humidity). For cooling applications,
this steep increase should ideally occur between 0.05 < P/P0 < 0.327 and the regeneration temperature
should be close to that of the cogeneration plants for building heating (≈65 °C).14 To date, the most
promising water‐stable MOFs displaying S‐shape isotherms in the chilling range are MIL‐125‐NH2,12 Al‐
fum,15 Zr‐fum,16 CAU‐10,17 MIP‐200,14 and Co2Cl2(BTDD).18 Table 1 summarizes values of some
performance parameters for these MOFs when used as cooling adsorbents.

- Insert Table 1 –

A principal concern with use of MOFs in water‐based applications is their long‐term stability,
suggesting the need for alternative porous materials.19 An interesting candidate is covalent‐organic
frameworks (COFs),20, 21 which have recently been proposed for water‐based applications, as they
exhibit high water uptake and S‐shape isotherms in the range of interest (0.05–0.32 P/P0),22 while
also showing high structural and chemical tunability and remaining relatively chemically stable to
water, to acid, and under redox conditions.23-28 A particularly promising COF for water‐based
applications is TpPa‐1.29 First reported by Banerjee and co‐workers in 2012,30 TpPa‐1 is a 2D imine‐
based COF that comprises 1,3,5‐triformylphloroglucinol (Tp) and p‐phenylenediamine (Pa‐1)
(Figure 1a), exhibits a water uptake of 30 wt% at 0.3 P/P0,29 is highly stable in water, and is bulk‐scale
processable.31 Accordingly, it was described to be an excellent desiccant under ambient conditions.

-Insert figure 1-

Herein, we report that TpPa‐1 can serve as an efficient adsorbent for cooling systems in cogeneration
plants that typically work around 65 °C. We show that TpPa‐1 can operate at a regeneration
temperature below 65 °C (temperature needed for complete water desorption). In tests mimicking
standard operating conditions for industry, its cooling coefficient of performance (COPC) was 0.77.
Finally, we demonstrate that visible light can be used to regenerate TpPa‐1 (i.e., remove all its water
molecules) and consequently, to drive the cooling process.

TpPa‐1 was synthesized following the procedure of Banerjee and co‐workers.29 A mixture of Tp and
Pa‐1 (1:1.5 molar ratio) was dissolved in a mixture of 1,4‐dioxane and mesitylene, and the resulting
solution was mixed with acetic acid. Then, this solution was heated at 120 °C for 3 days in a sealed
glass reactor, ultimately yielding a bright red solid, which was gently washed with N,N‐
dimethylacetamide, acetone, and tetrahydrofuran. X‐ray powder diffraction (XRPD) of the resultant
powder after overnight drying (dynamic vacuum; 80 °C) indicated formation of the expected eclipsed
crystalline phase of TpPa‐1 (Figure 1b). Nitrogen physical‐adsorption measurements on a sample
of TpPa‐1 outgassed at 120 °C followed a pure type I isotherm,32 and the calculated Brunauer–
Emmett–Teller (BET) surface area was 840 m2 g−1 (Figure S1, Supporting Information).29 The pore
volume was also accessible to CO2, with a total amount of 11.1 mmol g−1 at 760 Torr (Figure S2,
Supporting Information). The collected water isotherm revealed an “S‐”type sorption isotherm,
exhibiting a main, steep uptake at α = 0.22 and a maximum uptake of 0.45 gwater gCOF−1 at
0.90 P/P0 (Figure 1c). This S‐shape trend is associated with an initial sorption of water molecules
around the polar groups of the framework via hydrogen bonding, which leads to gradual formation of
a monolayer on the inner surface of the pore before the hole cavity becomes filled.33 Importantly, all
these sorption data are fully consistent with previously reported values for this COF.29

The energy efficiency of porous materials for cooling applications is commonly determined by their
COPC,7 which is defined as the ratio of vaporization heat (Qev) to regeneration heat (Qreg); in other
words: the useful output energy obtained in function of the energy demand (Equations (S1)–(S5) and
Figure S3, Supporting Information). To evaluate both Qev and Qreg, we analyzed an isosteric cycle
diagram of an adsorption air‐conditioning cycle to determine the working capacity (Δw) of the working
pair TpPa‐1/H2O and the desorption temperature (Tdes) (Figure 2a). This diagram was calculated using
the water adsorption isobars in a range of water‐vapor pressure values (0.7, 1.2, 2.4, 3.7, and 5.6 kPa)
and under variable Tdes (Figure S4, Supporting Information), with the operational temperature of the
cycle evaporation (Tev) fixed at 10 °C, and the temperature of adsorption and condensation (Tad = Tcon),
at 30 °C. During isobaric adsorption (step IV–I), TpPa‐1 adsorbed water, reaching a maximum uptake
of 0.27 gwater gCOF−1. Then, during isosteric heating (I–II), TpPa‐1 became fully saturated and the
pressure increased from 1.2 to 4.2 kPa by increasing the temperature from 30 to 37.2 °C without
desorption. During isobaric desorption (II–III), heating continued and desorption proceeded until
a Tdes of 65 °C was reached, at which point the water uptake was minimal. Finally, during isosteric
cooling, decreasing the temperature, the pressure was reduced and TpPa‐1 was regenerated. Thus,
the working capacity depends on Tdes and increases up to 0.27 gwater gCOF−1 at Tdes = 65 °C or higher.

-insert figure 2-

Next, we determined the heat of adsorption (ΔadsH), by using water isotherms collected at two
temperatures (25 and 40 °C; Figure 1c and Figure S5 (Supporting Information)) and by adjusting the
obtained values to the Clausius–Clapeyron equation (Equation (S6), Supporting Information). After a
high‐energy interaction at low coverage (70 kJ mol−1), the ΔadsH decayed down to 45–50 kJ mol−1,
showing a value closer to that of the enthalpy of evaporation of water (Figure 1d).34 This low value
was attributed to water–water molecular interactions, which are more favorable than COF–water
interactions. Note that our determined ΔadsH as a function of the water uptake closely agrees with a
previously reported value from a simulation.33

Having determined Δw, Tdes, and ΔadsH for TpPa‐1, we then calculated its COPC as a function
of Tdes (Figure 2b). A value of 0.77 was found in the range from 45 to 65 °C. Remarkably, this value falls
within the range of reported values for the three highest‐performing MOFs (MIP‐200, Co2Cl2[BTDD],
and CAU‐10), which can perform at full efficiency in building–heating cogeneration plants (63 °C). Note
that we also performed 40 consecutive water adsorption and desorption cycles under near‐
operational conditions for air‐conditioning systems (P = 2.36 kPa, Tads = 303 K, and Tdes = 383 K; Figure
S6, Supporting Information). This cycling experiment corroborated that TpPa‐1 exhibits a high degree
of stability (Figure S7, Supporting Information), with less than 0.02 gwater gCOF−1 loss of uptake after
cycling.

Solar collectors are a green alternative to provide the heat required for operation of AHTS systems.
For example, they are commonly in demand for nonelectric, water‐based refrigerators used for long‐
term storage of harvested grains; especially for grain silos located in areas with limited access to
electricity. Inspired by this precedent and by the photothermal effect that COF exhibits during
irradiation (Figure S8, Supporting Information),35 we explored the use of TpPa‐1 as an adsorbent for
photoactivated desorption of water (i.e., photoactivated cooling). To this end, TpPa‐1 was exposed to
visible light at an irradiance of 0.32 mW cm−2. Under these conditions, TpPa‐1 immediately heated up,
reaching 65 °C in less than 1 min (Figure 3a,c). Once we confirmed that the working desorption
temperature could be reached by light irradiation, we then studied the light‐triggered release of water
in TpPa‐1 by coupling the radiation source to the sorption instrument (Figure 3b). At an operating
water pressure of 2.36 kPa, TpPa‐1 initially adsorbed 0.25 gwater gCOF−1. Once the light had been
switched on, the water was fully desorbed from TpPa‐1. Finally, the on/off light‐switching cycle was
repeated to perform 15 consecutive adsorption–desorption cycles (Figure 3d), in which no loss of
performance was detected. These findings together demonstrated that TpPa‐1 could be efficiently
regenerated by using visible light.
-insert figure 3-

In conclusion, and to the best of our knowledge, we have reported the first‐ever use of a COF (TpPa‐
1) as an adsorbent in adsorption‐based cooling processes. Relative to other MOF‐, zeolite‐, and silica‐
based adsorbents previously reported for this application, TpPa‐1 exhibits comparable working
capacity, a lower energy of adsorption, and a lower regeneration temperature (Table 1).12, 14-16, 18, 36-
42
   Accordingly, TpPa‐1 ranks alongside the highest‐performing MOFs tested to date, showing its full
efficiency below 65 °C. Furthermore, we have exploited the photothermal effect of TpPa‐1 in the
visible region to enable light‐triggered water desorption, thus paving the way to use of TpPa‐1 in solar‐
driven cooling. We are confident that the possibility of synthesizing COFs with other chemically stable
linkages, such as amine, amide, alkene, and polyarylether,43, 44 could pave the way to new functional
adsorbents for this promising technology to reduce society's electrical energy demands.

Experimental Section
Synthesis of TpPa‐1: Tp (0.3 mmol; 63 mg) and Pa‐1 (0.45 mmol; 49 mg) were separately dissolved in
a 1:1 mixture of mesitylene and 1,4‐dioxane (1.5 mL each) and mixed together with 0.5 mL of
8 M acetic acid inside a Pyrex vial (13 mL). The mixture was sonicated for 5 min to assure complete
homogenization and then, was heated at 120 °C for 3 days. The resultant bright red solid was filtered
off and washed twice with N,N‐dimethylacetamide (10 mL), twice with acetone (10 mL), and once with
tetrahydrofuran. The resultant powder was dried under dynamic vacuum at 80 °C overnight.

Methods: XRPD patterns were collected on an X'Pert PRO MPDP analytical diffractometer (Panalytical)
at 45 kV, 40 mA using Cu Kα radiation (λ = 1.5419 Å). Volumetric N2 adsorption–desorption isotherm
was collected at 77 K using an ASAP2020 HD (Micromeritics). BET surface area was determined using
Microactive software (Micromeritics). Volumetric CO2 adsorption–desorption isotherm was collected
at 203 K using an ASAP2020 HD (Micromeritics) coupled with a chiller. Gravimetric water vapor
adsorption–desorption isotherms were measured using a dynamic vapor sorption (DVS) vacuum
instrument (Surface Measurement Systems Ltd.). The weight of the dried powder (≈20 mg) was
constantly monitored with a high‐resolution microbalance (±0.1 µg) and recorded at 25 and 40 °C (±0.2
°C) under pure water‐vapor pressures. The kinetics curves of water‐vapor adsorption were obtained
measuring real‐time mass change. The isobars were recorded at different temperatures (range: 110–
30 °C) at the fixed pressures of 0.7, 1.2, 2.4, 3.7, and 5.6 kPa. Prior to the sorption experiments,
samples were degassed inside the chamber under vacuum at 120 °C for 6 h. The heat‐capacity
measurements were performed on a differential scanning calorimeter (Mettler Toledo). The heating
rate used was 10 °C min−1 (range: 10–90 °C) and sapphire was used as a reference material. Visible
irradiation was supplied by a Bluepoint 4 Ecocure (Hönle UV Technology) intensity spot lamp (300–
650 nm) after a UV filter (300–400 nm). Temperature was recorded in infrared camera PI 450 (Optris),
working in the range of 0–250 °C. Data were obtained using the PI Connect software. Light‐triggered
kinetic water vapor adsorption curves were collected measuring real‐time mass change under
constant pressure of pure water and by switching the lamp on and off for irradiation cycling. For this
application, a DVS vacuum instrument was used as follows (see Figure 3b): by including a glass
viewport below the chamber to enable passage of visible light; by using a flat quartz pan; and by
aligning the lamp and the pan with a center plug. Solid‐state diffusive refraction spectra of the pure
powder were recorded using a Cary 4000 spectrophotometer (Agilent Technologies) in the wavelength
range of 300–900 nm.

Acknowledgements
This work was supported by the Spanish MINECO (project PN MAT2015‐65354‐C2‐1‐R); the Catalan
AGAUR (project 2017 SGR 328); the ERC, under EU‐FP7 (ERC‐Co 615954); and the CERCA
Program/Generalitat de Catalunya. ICN2 was supported by the Severo Ochoa program from Spanish
MINECO (Grant No. SEV‐2017‐0706).

Conflict of Interest
The authors declare no conflict of interest.

Keywords
adsorption heat transformation, cooling, covalent-organic framework,

porous materials, visible light

Received: May 10, 2019

Revised: August 7, 2019

Published online: September 3, 2019

References
[1] H. Kummer, F. Jeremias, A. Warlo, G. Fuldner, D. Frohlich, C. Janiak, R. Glaser, S. K. Henninger, Ind.
Eng. Chem. Res. 2017, 56, 8393.

[2] S. Narayanan, X. Li, S. Yang, H. Kim, A. Umans, I. S. McKay, E. N. Wang, Appl. Energy 2015, 149, 104.

[3] J. Canivet, A. Fateeva, Y. Guo, B. Coasne, D. Farrusseng, Chem. Soc. Rev. 2014, 43, 5594.

[4] S. K. Henninger, F. P. Schmidt, H. M. Henning, Adsorption 2011, 17, 833.

[5] L. G. Gordeeva, Y. I. Aristov, Energy 2019, 167, 440.

[6] I. S. Girnik, Energy 2016, 106, 13.

[7] M. F. de Lange, K. J. F. M. Verouden, T. J. H. Vlugt, J. Gascon, F. Kapteijn, Chem. Rev. 2015, 115,
12205.

[8] W. Wang, L. Wu, Z. Li, Y. Fang, J. Ding, J. Xiao, Drying Technol. 2013, 31, 1334.

[9] X. Wei, W. Wang, J. Xiao, L. Zhang, H. Chen, J. Ding, Chem. Eng. J. 2013, 228, 1133.

[10] P. Goyal, P. Baredar, A. Mittal, A. R. Siddiqui, Renewable Sustainable Energy Rev. 2016, 53, 1389.

[11] J. Canivet, J. Bonnefoy, C. Daniel, A. Legrand, B. Coasne, D. Farrusseng, New J. Chem. 2014, 38,
3102.

[12] L. G. Gordeeva, M. V. Solovyeva, Y. I. Aristov, Energy 2016, 100, 18.

[13] F. Jeremias, D. Frohlich, C. Janiak, S. K. Henninger, New J. Chem. 2014, 38, 1846.

[14] S. Wang, J. S. Lee, M. Wahiduzzaman, J. Park, M. Muschi, C. Martineau-Corcos, A. Tissot, K. H.
Cho, J. Marrot, W. Shepard, G. Maurin, J.-S. Chang, C. Serre, Nat. Energy 2018, 3, 985.

[15] F. Jeremias, D. Frohlich, C. Janiak, S. K. Henninger, RSC Adv. 2014, 4, 24073.

[16] M. V. Solovyeva, L. G. Gordeeva, T. A. Krieger, Y. I. Aristov, Energy Convers. Manage. 2018, 174,
356.
[17] D. Frohlich, E. Pantatosaki, P. D. Kolokathis, K. Markey, H. Reinsch, M. Baumgartner, M. A. van der
Veen, D. E. De Vos, N. Stock, G. K. Papadopoulos, S. K. Henninger, C. Janiak, J. Mater. Chem. A 2016, 4,
11859.

[18] A. J. Rieth, S. Yang, E. N. Wang, M. Dinca˘, ACS Cent. Sci. 2017, 3, 668.

[19] N. C. Burtch, H. Jasuja, K. S. Walton, Chem. Rev. 2014, 114, 10575.

[20] A. P. Cote, A. I. Benin, N. W. Ockwig, M. O’Keeffe, A. J. Matzger, O. M. Yaghi, Science 2005, 310,
1166.

[21] Rebecca L. Li, N. C. Flanders, A. M. Evans, W. Ji, I. Castano, L. X. Chen, N. C. Gianneschi, W. R.
Dichtel, Chem. Sci. 2019, 10, 3796.

[22] L. Stegbauer, M. W. Hahn, A. Jentys, G. Savasci, C. Ochsenfeld, J. A. Lercher, B. V. Lotsch, Chem.
Mater. 2015, 27, 7874.

[23] N. Huang, P. Wang, D. Jiang, Nat. Rev. Mater. 2016, 1, 16068.

[24] X. Li, C. Zhang, S. Cai, X. Lei, V. Altoe, F. Hong, J. J. Urban, J. Ciston, E. M. Chan, Y. Liu, Nat. Commun.
2018, 9, 2998.

[25] T. Sick, A. G. Hufnagel, J. Kampmann, I. Kondofersky, M. Calik, J. M. Rotter, A. Evans, M. Doblinger,
S. Herbert, K. Peters, D. Bohm,

P. Knochel, D. D. Medina, D. Fattakhova-Rohlfing, T. Bein, J. Am. Chem. Soc. 2018, 140, 2085.

[26] P. Pachfule, A. Acharjya, J. Roeser, T. Langenhahn, M. Schwarze, R. Schomacker, A. Thomas, J.
Schmidt, J. Am. Chem. Soc. 2018, 140, 1423.

[27] L. Stegbauer, K. Schwinghammer, B. V. Lotsch, Chem. Sci. 2014, 5, 2789.

[28] J. Xie, S. A. Shevlin, Q. Ruan, S. J. A. Moniz, Y. Liu, X. Liu, Y. Li, C. C. Lau, Z. X. Guo, J. Tang, Energy
Environ. Sci. 2018, 11, 1617.

[29] B. P. Biswal, S. Kandambeth, S. Chandra, D. B. Shinde, S. Bera, S. Karak, B. Garai, U. K. Kharul, R.
Banerjee, J. Mater. Chem. A 2015, 3, 23664.

[30] S. Kandambeth, A. Mallick, B. Lukose, M. V. Mane, T. Heine, R. Banerjee, J. Am. Chem. Soc. 2012,
134, 19524.

[31] S. Karak, S. Kandambeth, B. P. Biswal, H. S. Sasmal, S. Kumar, P. Pachfule, R. Banerjee, J. Am.
Chem. Soc. 2017, 139, 1856.

[32] K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol, T. Siemienewska,
Pure Appl. Chem. 1985, 57, 603.

[33] Y. Ge, H. Zhou, Y. Ji, L. Ding, Y. Cheng, R. Wang, S. Yang, Y. Liu, X. Wu, Y. Li, J. Phys. Chem. C 2018,
122, 27495.

[34] L. Garzon-Tovar, J. Perez-Carvajal, I. Imaz, D. Maspoch, Adv. Funct. Mater. 2017, 27, 1606424.

[35] J. Espin, L. Garzon-Tovar, A. Carne-Sanchez, I. Imaz, D. Maspoch, ACS Appl. Mater. Interfaces 2018,
10, 9555.

[36] B. Shi, R. Al-Dadah, S. Mahmoud, A. Elsayed, Appl. Therm. Eng. 2016, 106, 325.
[37] D. C. Wang, Z. Z. Xia, J. Y. Wu, R. Z. Wang, H. Zhai, W. D. Dou, Int. J. Refrig. 2005, 28, 1073.

[38] C. Y. Tso, C. Y. H. Chao, Int. J. Refrig. 2012, 35, 1626.

[39] L. Z. Zhang, Appl. Therm. Eng. 2000, 20, 103.

[40] F. B. Cortes, F. Chejne, F. Carrasco-Marin, C. Moreno-Castilla, A. F. Perez-Cadenas, Adsorption
2010, 16, 141.

[41] S. Vasta, A. Freni, A. Sapienza, F. Costa, G. Restuccia, Int. J. Refrig. 2012, 35, 701.

[42] M. Fischer, Phys. Chem. Chem. Phys. 2015, 17, 25260.

[43] S. J. Lyle, P. J. Waller, O. M. Yaghi, Trends Chem. 2019, 1, 172.

[44] X. Guan, H. Li, Y. Ma, M. Xue, Q. Fang, Y. Yan, V. Valtchev, S. Qiu, Nat. Chem. 2019, 11, 587.
Table 1. Comparison of adsorption performance parameters for highperforming MOFs (literature
values) and TpPa-1 (experimental values from this study): water loading, regeneration temperature
(Tdes), COPC, and heat of adsorption (ΔadsH).
Figure 1. a) The chemical structure of TpPa-1 (color code: C, gray; N, blue; O, red; H atoms have been
omitted for clarity). b) XRPD on the simulated structure (red) and on the experimental data record
(black). c) Water adsorption–desorption isotherm at 25 °C. d) Heat of adsorption as a function of
loading. Solid dots: adsorption branches; hollow dots: desorption branches.
Figure 2. a) Isosteric cycle diagram for the working pair TpPa-1/water, calculated for an air-cooling
cycle. b) COPC as a function of Tdes for the working pair TpPa-1/water.
Figure 3. a) IR camera images of TpPa-1 (top); after 60 s of light irradiation (middle); and after ≈60 s
after the light had been switched off (bottom). b) Photograph, and the corresponding scheme of the
light coupled to the sorption instrument. c) Temperature cycles induced by switching the light on and
off. d) Water adsorption–desorption cycles induced by switching the light on and off.