Modulation of Turing Patterns in the CDIMA Reaction by Ultraviolet

Subscriber access provided by Columbia University Libraries

A: Kinetics, Dynamics, Photochemistry, and Excited States

Modulation of Turing Patterns in the CDIMA Reaction by Ultraviolet and Visible Light Prof. Dr. Raphael Nagao, Renan Carneiro Cavalcante de Miranda, Irving R. Epstein, and Milos Dolnik J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.8b10819 • Publication Date (Web): 15 Jan 2019 Downloaded from http://pubs.acs.org on January 16, 2019

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 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 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.

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

Modulation of Turing Patterns in the CDIMA Reaction by Ultraviolet and Visible Light

Raphael Nagao1,2,*, Renan C. C. de Miranda1, Irving R. Epstein3, Milos Dolnik3,*

1Institute 2Center

of Chemistry, University of Campinas, CEP 13083-970 Campinas, SP, Brazil

for Innovation on New Energies, University of Campinas, CEP 13083-841 Campinas, SP,

Brazil 3Department

of Chemistry and Volen Center for Complex Systems, Brandeis University, Waltham MA 02454, USA

Abstract We have carried out the first systematic study of the effects of ultraviolet light, both alone and in combination with visible white light, on Turing patterns in the chlorine dioxide-iodine-malonic acid (CDIMA) reaction. The ultraviolet light used has a sharp peak at 368 nm and can perturb the system selectively. It primarily decomposes chlorine dioxide in a zeroth-order reaction and, when it is used to illuminate Turing patterns, shrunken spots are formed with an imperfect hexagonal arrangement. The ultraviolet light competes directly with the visible white light via the photoreaction with dissolved chlorine dioxide, which prevents the total suppression of patterns at intermediate intensities of white light. These results suggest that specific wavelengths of light in the ultraviolet spectrum selectively modify the chemistry behind the pattern formation and can be utilized to generate novel self-organized structures under forcing conditions. We propose a modified Lengyel-Epstein model to incorporate the effect of ultraviolet illumination and obtain good qualitative agreement between simulations and experiments. These results support the idea that chlorine dioxide photoreaction is a key step in modulating CDIMA patterns under ultraviolet illumination.

*corresponding authors: [email protected] and [email protected]

1 ACS Paragon Plus Environment

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

1. Introduction Turing patterns (TP)1 have been thoroughly studied in the chlorine dioxide–iodine–malonic acid (CDIMA) reaction-diffusion system for over two decades2-5. The photosensitivity of this reaction is conveniently used to control and modify the pattern formation6. Most of these studies have employed visible white light (VWL) with weak and intermediate intensities7-12. Although it is well understood that the absorption of photons by molecular iodine and its photodissociation to very reactive atomic iodine trigger the system photosensitivity, the details of the photochemistry in this system are not completely understood13-15. In addition to molecular iodine, there are other species present, both initial reagents and intermediates that can absorb light in the visible spectrum and thus may contribute to the photosensitivity of the reaction. Weak and intermediate illumination by VWL results in the consumption of iodide ions in the system, which can be monitored experimentally via the colored complex, CI3-, where C stands for a complex molecule (starch or poly-(vinyl alcohol))16. Accordingly, the final state is moved to a brighter domain due to the decrease of [CI3-]13. Experiments with strong VWL illumination, on the other hand, reveal that light may also induce the production of iodide under certain conditions, leading to a dark state17-18. Although much attention has been devoted to the control of TP by VWL, no previous study has considered more energetic wavelengths to perturb TP. In fact, ultraviolet light (UVL) can be effectively used to study the modulation of TP as well. Since molecular iodine absorbs light in the visible range with a maximum absorbance near 460 nm19, ultraviolet illumination is not expected to have the same strong effect on the photodissociation of molecular iodine and, therefore, different effects may be expected if UVL is used instead of VWL in experiments. In this study, we present the results of illumination of TP under ultraviolet and visible white light, either separately or in combination. Overall, UVL promotes the formation of shrunken spots with an irregular arrangement in space and competes with VWL for chlorine dioxide, allowing patterns to persist at intermediate VWL intensities that would suppress patterns in the absence of UVL. Furthermore, our experiments suggest that UVL contributes mainly to the decomposition of chlorine dioxide and indirectly affects the concentrations of chlorite and iodide in the system, while weak or intermediate VWL illumination leads primarily to consumption of iodide ions.


Page 2 of 16

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


2. Experimental Procedure The initial concentrations of the substances at the instant of mixing in the feeding chamber of the continuously fed stirred tank reactor (CSTR) were: [I2]0 = 0.4 mM (iodine, Sigma-Aldrich), [MA]0 = 1.8 mM (malonic acid, Sigma-Aldrich) and [PVA]0 = 10 g L-1 (poly-(vinyl alcohol) with average molecular weight of 9,000-10,000 and 80% hydrolyzed, Sigma-Aldrich). Chlorine dioxide, prepared as described in ref. 20, was utilized at two different concentrations: [ClO2]0 = 0.05 mM and [ClO2]0 = 0.14 mM. All solutions contained 10 mM H2SO4 (sulfuric acid, Fisher). The CSTR was fed independently at the same flow rate through three peristaltic pumps (Rainin). The residence time was 160 s, and the homogeneity of the reaction mixture was maintained by three magnetic stirring bars with a constant rotation rate of 1000 rpm. The CSTR served as a feeding chamber for the continuously-fed unstirred reactor (CFUR) and the reactors were separated by two membranes: a cellulose nitrate membrane with pore size 0.45 μm and thickness 0.12 mm (Whatman) placed beneath the gel and an anopore membrane with pore size 0.2 μm and thickness 0.10 mm (Whatman) impregnated with 4 % agarose gel. The reactor assembly was thermostated at 4 °C. The CFUR consisted of a 2 % agarose gel with diameter 25 mm and thickness 0.3 mm (Sigma-Aldrich). It was placed on one side against an optical glass window through which the illumination was applied, and on the other side it was coupled with a CSTR. A detailed description of the experimental apparatus is reported in ref. 21. A DLP projector 1510X (Dell) was used to implement uniform visible white light illumination of the CFUR, while ultraviolet light, with a peak centered at 368 nm, was applied by a controlling module HLV2 series (CCS). The Turing patterns were illuminated for 2 h. A CCD camera (Pulnix) equipped with a camera control unit (Hamamatsu) was used to record the images. Snapshots were taken in ambient light of intensity 0.6 mW cm-2 with no illumination projected on the CFUR. 3. Experimental Results Before starting the investigation of the effect of VWL and UVL on TP, the emission spectra of both illuminations were recorded, see Figure 1. The measurements were carried out in the interval 210-860 nm, and the relative emission intensities for VWL (full-line) and UVL (dashed-



line) are shown in Figure 1a. As observed, our visible light source presents a spectrum in the range of 400 to 700 nm with 6 prominent peaks located at 434, 441, 487, 512, 551 and 577 nm. These peaks can be grouped in three pairs, i.e. (434, 441) nm, (487, 512) nm and (551, 577) nm. The wavelengths in these domains correspond well with the RGB color scale commonly used in commercially available projectors (our source of visible white light). This type of illumination has been used to control TP with different spatial perturbations8-12. On the other hand, the UVL (dashed-line) shows single sharp peak at 368 nm from the UV LED lamps that were utilized in the experiments.

Figure 1. (a) VWL (full line) and UVL (dashed line) emission spectra; (b) iodine (full line), chlorine dioxide (dashed line) and triiodide (dotted line) absorption spectra. Based on these observations, we undertook a detailed investigation of possible candidates, i.e. reactants or intermediates in the CDIMA reaction, sensitive to light perturbation. It has been 4 ACS Paragon Plus Environment

Page 4 of 16

Page 5 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


demonstrated that VWL affects the reaction that produces I· radicals from the bond-breaking of iodine molecules13. As shown in Figure 1b, the absorption spectrum of I2 (full line) has a maximum at 460 nm, which demonstrates the susceptibility of the reaction to VWL illumination. This photochemical reaction is the rate-determining step and triggers several subsequent reactions that occur in consecutive and parallel pathways. Overall, the effect of VWL in the CDIMA reaction is summarized by the global step r1 proposed by Muñuzuri et al.13, hν(VWL) + 2ClO2 + 2I- → 2ClO2- + I2

(r1)

In other words, VWL consumes chlorine dioxide and iodide, forming chlorite and iodine. Several species are affected by UVL: chlorine dioxide (ClO2), chlorite (ClO2-), chlorous acid (HClO2), hypochlorous acid (HClO), iodine (I2), triiodide (I3-), iodous acid (HIO2) and hypoiodous acid (HOI). The CDIMA reaction is quite complex, with an intricate mechanism and multiple intermediates. Our analysis was facilitated by an extensive study done by Li and coworkers on the spectroscopic characterization of the substances present in CDIMA reaction19. ClO2-, HClO2, HClO, HIO2 are sensitive to UVL but are not strongly affected at 368 nm, while I2 and HOI have a small absorbance in this region. However, two species have a significant absorbance at 368 nm: ClO2 (dashed-line) centered at 351 nm and I3- (dotted-line) with peaks at 280 and 350 nm as shown in Figure 1b. The wavelengths in the ultraviolet range have enough energy to decompose triiodide and chlorine dioxide, but no effect on triiodide has been observed when it is the only species in solution. A rapid recombination restores the molecule; a similar phenomenon occurs in solutions containing only iodine. When a mixture of I3- and ClO2 is illuminated with VWL and UVL, no significant change is observed at 280 nm, which indicates a coadjutant role for I3- during ultraviolet illumination (data not shown). The photochemical decomposition of chlorine dioxide by UVL, with zeroth-order kinetics, is a well-known reaction, and several mechanisms have been proposed22-24. Reaction r2 summarizes the mechanism proposed by Karpel Vel Leitner et al.22. hν(UVL) + 10ClO2 + 5H2O → 6ClO3- + 4Cl- + 7/2O2 + 10H+

(r2)

Reactions r1 and r2 thus indicate that VWL and UVL can compete for the consumption of chlorine dioxide if both are applied simultaneously. While VWL induces a bright state because of 5 ACS Paragon Plus Environment


the consumption of iodide species, and consequently, a decrease in the concentration of CI3(responsible for the contrast in the Turing patterns), UVL decomposes ClO2 and weakens the effect of the VWL. To test this hypothesis, we have applied both types of illumination, individually and in combination, to TP formed spontaneously in the CDIMA reaction at two distinct concentrations of chlorine dioxide (Figures 2 and 3). Figure 2a depicts a stationary TP consisting of a combination of stripes and spots with an intrinsic wavelength of λP = 0.63 ± 0.02 mm at [ClO2] = 0.05 mM (see the Experimental Procedure section for additional details). Lower ClO2 concentrations result in an increase of λP25. When VWL is applied to the patterns at an intensity of IVWL = 3 mW cm-2 (Figure 2b), the stripes and spots connect to form longer stripes. This is in agreement with reaction r1, in which higher intensities of VWL should consume iodide and produce chlorite, creating a brighter average state. At higher intensities, e.g., IVWL = 6 mW cm-2 (Figure 2c) and IVWL = 9 mW cm-2 (Figure 2d), TP are suppressed and a bright state is reached.

Figure 2. (a) Stationary Turing patterns (λP = 0.63 ± 0.02 mm) and under VWL illumination with (b) IVWL = 3 mW cm-2, (c) IVWL = 6 mW cm-2, (d) IVWL = 9 mW cm-2; under UVL illumination with (e) IUVL = 1.3 mW cm-2 and a combination of VWL and UVL with (f) IUVL = 1.3 mW cm-2 + IVWL = 3 mW cm-2, (g) IUVL = 1.3 mW cm-2 + IVWL = 6 mW cm-2, (h) IUVL = 1.3 mW cm-2 + IVWL = 9 mW cm-2. Each snapshot is 5×5 mm. [ClO2] = 0.05 mM. Figure 2e shows the effect of UVL at IUVL = 1.3 mW cm-2 on TP after 2 h of illumination. A spot pattern emerges but with some differences relative to the usual spot domain found in the CDIMA reaction25-26. The white spots when perturbed by UVL shrink and do not display a regular 6 ACS Paragon Plus Environment

Page 6 of 16

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


hexagonal arrangement. This feature may result from ClO2 decomposition occurring mostly at the edges of the spots. A combination of VWL and UVL was also examined with light intensities: IUVL = 1.3 mW cm-2 + IVWL = 3 mW cm-2 (Figure 2f), IUVL = 1.3 mW cm-2 + IVWL = 6 mW cm-2 (Figure 2g), IUVL = 1.3 mW cm-2 + IVWL = 9 mW cm-2 (Figure 2h). When UVL is applied simultaneously with VWL, the shift to the bright domain does not occur as easily as when only VWL is used. In fact, UVL prevents the formation of longer stripes by connection between spots and stripes (Figure 2f) unless a higher intensity of VWL is applied (Figure 2g). The resemblance between the patterns shown in Figures 2a and 2f illustrates this observation. Even under an illumination intensity as high as IVWL = 9 mW cm-2, the patterns are still not completely obliterated in the presence of UVL (Figure 2h), in contrast to the case shown in Figure 2d, where only VWL was utilized and total suppression of TP was observed. The competition for the consumption of ClO2 becomes evident when Figures 2f, g and h are compared to Figures 2b, c and d, respectively. If [ClO2] is increased to 0.14 mM, a different scenario is found (Figure 3). First, the labyrinthine TP exhibits an intrinsic wavelength of λP = 0.40 ± 0.03 mm, in agreement with previous studies11-12. When VWL is used with intensities of 3 and 6 mW cm-2 (Figures 3b and c, respectively), the TP are immediately affected, and a spot domain (Hπ) is maintained after 2 h of illumination. Note, however, that at IVWL = 6 mW cm-2 the spots are less evident and vanish at IVWL = 9 mW cm-2 (Figure 3d), where the patterns are suppressed and a bright state predominates.



Page 8 of 16

Figure 3. (a) Stationary Turing patterns (λP = 0.40 ± 0.03 mm) and under VWL illumination with (b) IVWL = 3 mW cm-2, (c) IVWL = 6 mW cm-2, (d) IVWL = 9 mW cm-2; under UVL illumination with (e) IUVL = 1.3 mW cm-2 and a combination of VWL and UVL with (f) IUVL = 1.3 mW cm-2 + IVWL = 3 mW cm2, (g) I -2 -2 -2 UVL = 1.3 mW cm + IVWL = 6 mW cm , (h) IUVL = 1.3 mW cm + IVWL = 9 mW cm-2. Each snapshot is 5×5 mm. [ClO2] = 0.14 mM. When ultraviolet light is applied at IUVL = 1.3 mW cm-2 (Figure 3e), TP are also suppressed and a bright state is reached after 2 h. This behavior, which resembles that seen with VWL alone, can be explained in two ways: (a) there is a small absorption by I2 at 368 nm, which allows UVL to perturb the system similarly to VWL as described by reaction r1; (b) the higher [ClO2] (e.g. 0.14 mM) mitigates the effect of ClO2 decomposition by UVL. Additionally, the state seen in Figure 3e is slightly darker than that in Figure 3d, where only VWL is used for illumination. Considering

that

UVL

decomposes

chlorine

dioxide

photochemically,

lower

concentrations of this species may induce a darker state, as previously reported17. Figures 3f, g and h show the suppression of TP when a combination of VWL and UVL is applied. In all cases, a bright state is observed, which can be attributed to the consumption of I- via reaction r1 without any restriction from ClO2 replenishment because of its higher concentration. 4. Modeling We employed the Lengyel-Epstein model2 with the influence of VWL incorporated as an additive term w11-12 and applied uniformly in space. The partial differential equations that describe the spontaneous formation of Turing patterns (TP) are


Page 9 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


u uv  a u 4  w   2u t 1 u2

(1)

   v uv    b u   w   d 2 v  2 t 1 u    

(2)

Both equations are written in dimensionless form, where u and v are the concentrations of I- and ClO2- ions, respectively. The VWL illumination term (w) corresponds to the photochemical reaction r1. Additionally, the model also contains dimensionless parameters a, b, d and σ that are kept constant during the numerical integration. In order to study the effect of UVL on the TP, we modified the two parameters that depend upon [ClO2]  k1a MA0 a  ClO   k 2 0 2 

 I 2 0   k  I  2 0  1b

   

 k3b I 2 0   b  ClO   k  2 0 2  

(3)

(4)

where k1a (6.2 × 10-4 s-1) and k1b (5.0 × 10-5 M) are the rate constants for the reaction between malonic acid and iodine, k2 (900 M-1 s-1) is the rate constant for the reaction between chlorine dioxide and iodide, and k3b (9.2 × 10-5 s-1) is the rate constant for the reaction between chlorite and iodide. [MA]0, [I2]0 and [ClO2]0 are the initial concentrations of malonic acid, iodine and chlorine dioxide, respectively. The parameter α (~5.0 × 10-13) is an ad hoc constant that acts as a cutoff value when the concentration of iodide is very low. Additional details about the reactions and numerical values for the experimental parameters can be found in refs. 2 and 25. A key observation is the inverse relation between a and b and the initial concentration of chlorine dioxide. We incorporate the decomposition of ClO2 by UVL into the model via the parameters a’ and b’,

a' 

a 1 q


(5)


b' 

b 1 q

Page 10 of 16

(6)

where q is proportional to the UVL intensity. Based on the photochemical reaction r2 and the observation that a and b are inversely proportional to [ClO2]0, the decomposition of unreacted chlorine dioxide produces an increase in the parameter values a’ and b’. We simulate this effect by the insertion of q in the model through equations 5 and 6, so that an increase in UVL intensity results in an increase in a’ and b’. The partial differential equations that describe the photosensitivity of the TP to VWL (w) and UVL (q) are u uv  a 'u  4  w   2u t 1 u2

(7)

   v uv    b'  u   w   d 2 v  2 t 1 u    

(8)

The parameters used in our simulations are: (a) a = 12.4 and b = 0.19 (Figure 4); (b) a = 6.66 and b = 0.10 (Figure 5). The other parameters, d = 1.0 and σ = 49 were kept the same for both conditions. The chosen parameters correspond to [MA]0 = 2.0 mM, [I2]0 = 0.16 mM, (a) [ClO2]0 = 0.120 mM and (b) [ClO2]0 = 0.223 mM. The illumination terms have physical meaning only when 0 ≤ q < 1 and w ≥ 0. Our two dimensional simulations were calculated with 256×256 grid points separated by 0.5 space units with no-flux boundaries. The explicit Euler method with a time step of 0.001 and a 9-point stencil to compute the Laplacians was used. The initial conditions were set to the unstable steady state values of u (u* = a/5) and v (v* = 1 + (u*)2) with a small random fluctuation (1% of the steady state value) added to u at each grid point. 5. Numerical Results and Discussion We performed simulations of equations 7 and 8, as described in the previous section, in order to study the effect of VWL and UVL on TP. Figure 4 depicts the results in the interval of illumination 0 ≤ q ≤ 0.6 and 0 ≤ w ≤ 1.4.


Page 11 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


Figure 4. Snapshots of simulated TP under uniform VWL and UVL illumination. Fixed parameters: a = 12.4, b = 0.19, d = 1.0 and σ = 49. The snapshot is 84×84 space units (168×168 grid size). White color represents low concentration of variable u, black color represents high concentration of u. In the absence of illumination (w = 0, q = 0), labyrinthine TP are obtained. As expected, large values of a and b simulate lower concentrations of chlorine dioxide, setting the system close to the transition between TP and a steady state at a = 12.4, b = 0.19. In agreement with the experimental observations (Figures 2c and d), the simulations shown in Figure 4 result in complete suppression of TP at a single set of parameters (w = 1.4, q = 0). Random breaking of the labyrinthine patterns by UVL when w = 0 is observed. On the other hand, VWL induces a connection between spots and stripes by the production of chlorite (cf. reaction r1) as seen in the experiments (Figure 2b and g) and in the numerical calculations (Figure 4 at q = 0.6 in the interval 0 ≤ w ≤ 1.4). Overall, the simulations match well the experimental results. Remarkably, simulations of illumination utilizing both UWL and VWL produce a type of compensation effect like that observed in the experiments (Figures 2b, c and d compared to Figures 2f, g and h). In the presence of UVL, higher VWL intensities are required to move the system to a pattern similar to that obtained without UVL. For example, when q = 0.2 (Figure 4) the transition from labyrinthine to honeycomb patterns (H0) occurs at 0.6 ≤ w ≤ 1.0 instead of at 0.2 ≤ w ≤ 0.6 for q = 0. For q = 0.6, this transition could not be observed even at w = 1.4. This compensation



effect permeates the entire Figure 4. Comparing with the experiments, Figure 2d matches the qualitative behavior for w = 1.4, q = 0, and Figure 2h for w = 1.4, q = 0.2.

Figure 5. Snapshots of simulated TP under uniform VWL and UVL illuminations. Fixed parameters: a = 6.66, b = 0.10, d = 1.0 and σ = 49. The snapshot is 84×84 space units (168×168 grid size). We also simulated the modulation of TP by UVL and VWL at higher [ClO2] using the parameters a = 6.66, b = 0.10, d = 1.0 and σ = 49 (Figure 5). The first observation is the larger number of snapshots with a uniform state with lower u values (TP suppression), supporting the idea that ClO2 is a key species in the control of TP by UVL irradiation. The pattern suppression is found for w ≥ 0.2 at q = 0. In the experiments, we observed suppression of the patterns when IVWL ≥ 9 mW cm-2 (Figure 3d). When UVL illumination is applied in simulations (q = 0.2) the amplitude of the patterns increases and the TP are suppressed when w ≥ 0.6 (second row in Figure 5). When more intense UVL is used in simulations (q = 0.6), the intensity of VWL also has to be increased in order to suppress TP. In this case, as w increases, the transition to honeycomb pattern (H0) formation occurs in the range 1.0 ≤ w ≤ 1.4. Further increase of the light intensity results in compression of the black spots. This behavior was already found by Dolnik et al.7. In general, similar results as obtained for lower [ClO2] (Figure 4) were obtained in simulations for higher [ClO2] (Figure 5). The main difference is that smaller w (lower intensity 12 ACS Paragon Plus Environment

Page 12 of 16

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


VWL) is required to suppress TP in the case of higher [ClO2]. However, the model fails to reproduce the suppression of TP by only UVL as observed experimentally in Figure 3e. This is not surprising, since the model does not take into account I2 bond-breaking in the ultraviolet range, which would allow UVL to perturb the system in a manner similar to VWL as described by reaction r1. 6. Conclusions The photosensitivity of the CDIMA reaction to VWL is additive (w terms in equations (1) and (2)), corresponding to the photochemical reaction r1, in which I- is consumed and ClO2- is produced13. However, UVL does not affect this reaction in a similar way. Based on our experimental results, we can point out a close relation between the photochemical decomposition of chlorine dioxide and the ultraviolet spectrum. This is, indeed, a well-known reaction and some mechanisms have already been proposed22-24; see reaction r2 postulated by Karpel Vel Leitner et al.22. [ClO2] was varied experimentally from 0.05 to 0.14 mM, and distinct patterns were observed. First, UVL behaved similarly to VWL in suppressing TP and moving the system to a bright (low [I-]) state when an excess of [ClO2] was present. This observation can be explained by the small absorbance of I2 at 368 nm, which allows UVL to perturb the system similarly to VWL by breaking I2 molecules photochemically, which results in the global reaction r113. An alternative explanation might be that excess [ClO2] (0.14 mM) limits the effect of its decomposition by UVL. On the other hand, at lower [ClO2] (0.05 mM), the UVL-generated spots are slightly shrunken with an imperfect hexagonal arrangement due to random breaking of the stripes that constitute the labyrinthine TP. VWL seems to compete directly with UVL by the photoreaction with ClO2, requiring larger intensities to induce TP suppression. To study the combination between VWL and UVL on TP, we modified the Lengyel-Epstein model, introducing equations 5 and 6 to incorporate the inverse [ClO2] dependence of the dimensionless parameters a and b. The combination of the two light sources promoted the same compensation effect as observed experimentally, which supports our mechanistic interpretation. This systematic study opens the possibility to use different electromagnetic wavelengths, separately or combined, to control nonequilibrium pattern formation in the CDIMA reaction. 13 ACS Paragon Plus Environment


7. Acknowledgements R.N. (16/01817-9 and 16/22262-5) acknowledges the São Paulo State Foundation (FAPESP); R.C.C.M. acknowledges Coordination for the Improvement of Higher Education Personnel (CAPES) - Finance Code 001 for a scholarship. I.R.E. and M.D. acknowledge the National Science Foundation (CHE-1362477) for financial support. 8. References 1.

Turing, A. M. The Chemical Basis of Morphogenesis. Philos. Trans. R. Soc. Lond. Ser. BBiol. Sci. 1952, 237, 37-72.

2.

Lengyel, I.; Epstein, I. R. Modeling of Turing Structures in the Chlorite Iodide Malonic Acid Starch Reaction System. Science 1991, 251, 650-652.

3.

Kaen, M.; Satnoianu, R.; Muñuzuri, A. P.; Menzinger, M. Controlled Pattern Formation in the CDIMA Reaction with a Moving Boundary of Illumination. Phys. Chem. Chem. Phys. 2002, 4, 1315-1319.

4.

Haim, L.; Hagberg, A.; Nagao, R.; Steinberg, A. P.; Dolnik, M.; Epstein, I. R.; Meron, E. Fronts and Patterns in a Spatially Forced CDIMA Reaction. Phys. Chem. Chem. Phys. 2014, 16, 26137-26143.

5.

Gaskins, D. K.; Pruc, E. E.; Epstein, I. R.; Dolnik, M. Multifold Increases in Turing Pattern Wavelength in the Chlorine Dioxide-Iodine-Malonic Acid Reaction-Diffusion System. Phys. Rev. Lett. 2016, 117, 056001.

6.

Berenstein, I.; Yang, L.; Dolnik, M.; Zhabotinsky, A. M.; Epstein, I. R. Superlattice Turing Structures in a Photosensitive Reaction-Diffusion System. Phys. Rev. Lett. 2003, 91, 058302.

7.

Dolnik, M.; Zhabotinsky, A. M.; Epstein, I. R. Resonant Suppression of Turing Patterns by Periodic Illumination. Phys. Rev. E 2001, 63, 026101.

8.

Sagués, F.; Míguez, D. G.; Nicola, E. M.; Muñuzuri, A. P.; Casademunt, J.; Kramer, L. Travelling-Stripe Forcing of Turing Patterns. Physica D 2004, 199, 235-242.

9.

Míguez, D. G.; Pérez-Villar, V.; Muñuzuri, A. P. Turing Instability Controlled by Spatiotemporal Imposed Dynamics. Phys. Rev. E 2005, 71, 066217.

10.

Míguez, D. G.; Alonso, S.; Muñuzuri, A. P.; Sagués, F. Experimental Evidence of Localized Oscillations in the Photosensitive Chlorine Dioxide-Iodine-Malonic Acid Reaction. Phys. Rev. Lett. 2006, 97, 4. 14 ACS Paragon Plus Environment

Page 14 of 16

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


11.

Dolnik, M.; Bánsági, T., Jr.; Ansari, S.; Valent, I.; Epstein, I. R. Locking of Turing Patterns in the Chlorine Dioxide-Iodine-Malonic Acid Reaction with One-Dimensional Spatial Periodic Forcing. Phys. Chem. Chem. Phys. 2011, 13, 12578-12583.

12.

Feldman, D.; Nagao, R.; Bánsági, T., Jr.; Epstein, I. R.; Dolnik, M. Turing Patterns in the Chlorine Dioxide-Iodine-Malonic Acid Reaction with Square Spatial Periodic Forcing. Phys. Chem. Chem. Phys. 2012, 14, 6577-6583.

13.

Muñuzuri, A. P.; Dolnik, M.; Zhabotinsky, A. M.; Epstein, I. R. Control of the Chlorine Dioxide-Iodine-Malonic Acid Oscillating Reaction by Illumination. J. Am. Chem. Soc. 1999, 121, 8065-8069.

14.

Horvath, A. K.; Dolnik, M.; Zhabotinsky, A. M.; Epstein, I. R. Kinetics of Photoresponse of the Chlorine Dioxide-Iodine-Malonic Acid Reaction. J. Phys. Chem. A 2000, 104, 57665769.

15.

Rábai, G.; Kovács, K. M. Photoinduced Reaction between Chlorine Dioxide and Iodine in Acidic Aqueous Solution. J. Phys. Chem. A 2001, 105, 6167-6170.

16.

Epstein, I. R.; Pojman, J. A. Introduction to Nonlinear Chemical Dynamics. Oscillations, Waves, Patterns and Chaos. Oxford University Press: New York, 1998.

17.

Nagao, R.; Epstein, I. R.; Dolnik, M. Forcing of Turing Patterns in the Chlorine DioxideIodine-Malonic Acid Reaction with Strong Visible Light. J. Phys. Chem. A 2013, 117, 9120-9126.

18.

Steinberg, A. P.; Epstein, I. R.; Dolnik, M. Target Turing Patterns and Growth Dynamics in the Chlorine Dioxide–Iodine–Malonic Acid Reaction. J. Phys. Chem. A 2014, 118, 2393-2400.

19.

Li, J. Experimental and Mechanistic Studies of Chemical Reactions. Ph.D. Dissertation, Brandeis University, Waltham, MA, 1995.

20.

Brauer, G. Handbook of Preparative Inorganic Chemistry. Academic Press: New York, 1963.

21.

Nagao, R.; Varela, H. Turing Patterns in Chemical Systems. Quim. Nova 2016, 39, 474485.

22.

Karpel Vel Leitner, N.; De Laat, J.; Dore, M. Photodecomposition of Chlorine Dioxide and Chlorite by U.V. Irradiation - Part I. Photo-Products. Water Res. 1992, 26, 1655-1664.

23.

Cosson, H.; Ernst, W. R. Photodecomposition of Chlorine Dioxide and Sodium Chlorite in Aqueous Solution by Irradiation with Ultraviolet Light. Ind. Eng. Chem. Res. 1994, 33, 1468-1475.



24.

Quiroga, S. L.; Perissinotti, L. J. Reduced Mechanism for the 366 nm Chlorine Dioxide Photodecomposition in N2-Saturated Aqueous Solutions. J. Photoch. Photobio. A 2005, 171, 59-67.

25.

Rudovics, B.; Barillot, E.; Davies, P. W.; Dulos, E.; Boissonade, J.; De Kepper, P. Experimental Studies and Quantitative Modeling of Turing Patterns in the (Chlorine Dioxide, Iodine, Malonic Acid) Reaction. J. Phys. Chem. A 1999, 103, 1790-1800.

26.

Gunaratne, G. H.; Ouyang, Q.; Swinney, H. L. Pattern Formation in the Presence of Symmetries. Phys. Rev. E 1994, 50, 2802-2820.

TOC Graphic


Page 16 of 16

Modulation of Turing Patterns in the CDIMA Reaction by Ultraviolet

Recommend Documents