Colorimetric CO₂ Indicators

Conspectus

Carbon dioxide, CO₂, is an essential part of life, in that through green plant photosynthesis it is used to generate food and fuel and is generated in both aerobic and anaerobic respiration. Industrially, it is used in fire extinguishers, supercritical fluid extractions, and food packaging. Environmentally, it is in the atmosphere, hydrosphere, and biosphere and is responsible for global warming and the acidification of the oceans. The monitoring of CO₂ in the gas phase is usually carried out using FTIR spectroscopy, whereas the measurement of dissolved CO₂ usually involves an electrochemical device. Excitingly, the most recent forms of CO₂ indicators appear to offer significant advantages over current methods, such as simplicity, low cost, and portability.

This Account highlights the work of the Mills group on transforming CO₂ colorimetric indicator technology from the usual water-based (i.e., “wet”) indicator form to dry CO₂-sensitive inks, pigments, plastics, and adhesives. Initially, the basic theory associated with colorimetric CO₂ indicators is described, and the simple relationship between indicator absorbance and the partial pressure of CO₂, P_CO2, established. The early work on CO₂-sensitive inks is then described, where such inks comprise a hydrophilic pH-sensitive dye anion, coupled with a lipophilic quaternary ammonium cation, dissolved in a nonaqueous solution of a polymer which, when cast, forms a dry ink film that gives a reversible color response when exposed to CO₂ both in the gas phase and dissolved in solution. The ability to tune the sensitivity of a CO₂ ink film to the desired application through the judicious choice of the pH indicator dye and base concentration is described. The dependence of the sensitivity of a CO₂ ink film on temperature is used to create a temperature indicator, and the ability to tune the ink, to respond to high levels of CO₂, is used to create a fizziness indicator for carbonated drinks. Very sensitive CO₂ inks are used to make a vacuum and a general air-pressure indicator. The more recent development in CO₂ indicator technology is described in which CO₂ inks are used to coat silica particles to make a range of different CO₂-sensitive pigments, which, when incorporated into a plastic, through extrusion, produce a range of novel CO₂-sensitive plastic films that have many notable advantages over their ink film counterparts. Examples are then given of such plastic films being used for dissolved CO₂ measurements in salt water, for food packaging, and as an early wound-infection indicator. Finally, the recent incorporation of a CO₂-sensitive pigment into a pressure sensitive adhesive to make an after opening freshness tape is described briefly.

Although most commercial CO₂ indicators are assessed by eye and so are limited to qualitative analysis, this work shows that colorimetric CO₂ indicators can be used for quantitative analysis through absorbance measurements. Nowadays, such measurements can be readily made using just a digital camera and color analysis software via digital camera colorimetry, DCC, which is likely to have a significant impact on the widespread use of the CO₂ indicators described herein, their commercial viability, and their potential areas of application.

1. Introduction

Carbon dioxide, CO₂, is a basic chemical feedstock of life, as it is used to generate the fuel and food necessary for most life forms and is a common indicator of life and health. Although, there is little CO₂ in the atmosphere, ca. 412 ppmv (0.04%), it is rising and so creating environmental problems such as global warming and the acidification of the oceans.^1,2 Thus, the monitoring of the levels of CO₂ in the atmosphere, hydrosphere, and biosphere is a core part of environmental analysis.³ In industry, CO₂ is used as an inert gas in welding and fire extinguishers, as a pressurizing gas in oil recovery, and as a supercritical solvent in the decaffeination of coffee and supercritical drying.⁴ In the drinks industry, CO₂ is used to make a myriad of carbonated beverages and, in the food industry, as an active packaging gas, since it has antimicrobial activity. Its use in modified atmosphere packaging, MAP, accounts for over 60 billion food packages per annum.⁵ In medicine, the measurement of dissolved CO₂ levels in blood and the monitoring of the CO₂ in breath, i.e., capnography, are routine.^6,7

In recent years, a number of color-based CO₂ indicators have emerged as possible, inexpensive, disposable alternatives^8,9 to the usual bulky, expensive methods used to analyze for CO₂ in air or dissolved in solution, such as infrared spectroscopy and the Severinghaus electrode, respectively.^10,11 This Account outlines the recent evolution of CO₂ colorimetric indicators, from inks to pigments, plastic films, and adhesives, that has been pioneered by the Mills group. This is not to say other groups have not contributed to the development of such indicators,¹²⁻¹⁸ and a list of some of the major advances made by such groups over the years is given in Table S1 in the Supporting Information, SI, file that accompanies this paper.

There have been several major reviews on CO₂ indicators, and most have focused on their use in food packaging.¹⁹⁻²¹ However, this review is different in that, through the work of one group, the evolution of CO₂ indicators over the last 3 decades is described, with particular attention to how they work and potential applications other than in food packaging.

2. Optical CO₂ Indicators Theory

Most, if not all, color-based CO₂ indicators use a pH indicator dye, D, which responds to the change in pH in the surrounding (encapsulating) medium, be it an aqueous solution or an ink film, due to the formation of carbonic acid and its subsequent deprotonation. The latter process can be summarized by the following reaction,

1

In most, if not all, CO₂ indicators, sufficient base is present so that the partial pressure of CO₂, P_CO2, is related directly to the concentration of H⁺, [H⁺], i.e.,

2

where [Base] is the base concentration, which is usually sodium bicarbonate, NaHCO₃, in aqueous solution, and a quaternary ammonium hydroxide, Q⁺OH^–·xH₂O, in the case of CO₂-sensitive inks, pigments, and films. When a pH indicating dye is also present, as with a CO₂ colorimetric indicator, the following equilibrium is also set up,

3

where K_a is the acid dissociation constant of the pH indicator dye and HD and D^– are the protonated and deprotonated forms of the dye, with colors A and B, respectively. The dye concentration is very small and, usually, has no significant effect on the equilibrium reaction 1, and so the following expression can be derived, relating the color of the indicator to P_CO2,

4

where R is the ratio of the concentrations of HD and D^–, i.e., [HD]/[D^–], α is a proportionality constant (= K₁/K_a[Base]), and P_CO2 is the partial pressure of CO₂, which here, as is common practice, is expressed as a percentage of an atmosphere, where x% CO₂ = 0.01·x atm CO₂. Equation 4 is found to apply to most colorimetric CO₂ indicators. From eq 4 it follows that at R = 1, the indicator is halfway through its color change, and the associated level of P_CO2, P_CO2(S = 1/2), is equal to 1/α.

In order to use a CO₂ indicator for quantitative analysis, usually its absorbance, A, is monitored at a wavelength where D^– absorbs strongly, and in this work, in all the examples cited, A is the absorbance of the indicator at the maximum absorbance wavelength associated with the deprotonated form of the dye, D^–. Under such conditions, A is related to R, and so P_CO2 by the following expression,

5

where A_o and A_∞ are the fixed, measured absorbances of the CO₂ indicator when the dye is completely in its deprotonated and protonated form, respectively, i.e., in its extreme color forms, color B and color A, respectively. In practice, the values of A_o and A_∞ are usually taken as the measured values of A when the indicator is exposed to the extreme levels of CO₂, of 0 and 1 atm, respectively.

3. CO₂ Indicator Ink Films

Up until the early 1990s, work on CO₂ indicators had been limited to systems in which a pH indicator dye was dissolved in an aqueous solution. Examples include the drop-checker CO₂ indicator used in aquaria^22,23 and the “Einstein” indicator for ensuring correct tracheal intubation.¹² When such an indicator is used to measure the level of dissolved CO₂ in a test medium, the CO₂-sensitive aqueous indicator layer has to be confined behind a thin polymer film, such as polyethylene terephthalate, PTFE, which acts as a waterproof, gas-permeable (ion-impermeable) membrane, i.e., a GPM.^14,24

In the early 1990s, the Mills group were the first to report a “dry” CO₂ indicator film, created using a solvent-based ink^25,26 in which the highly hydrophilic anionic form of a pH indicator dye was rendered solvent-soluble by pairing it with the quaternary cation, Q⁺, of a phase transfer agent, PTA, Q⁺OH^–·xH₂O,

6

where the ion pair, Q⁺D^–·xH₂O, had the added attractive feature of possessing a few molecules of water of hydration even in a nonaqueous, lipophilic medium.²⁷ The above reaction allowed D^– to be dissolved in the same lipophilic solvent as a hydrophobic polymer, such as ethyl cellulose, EC, to form a solvent-based ink. On casting the ink onto an inert substrate and allowing it to dry, a thin, hydrophobic, water insoluble, CO₂ indicator ink film is created, which gives a reversible color response to CO₂,

7

where Q⁺D^–·xH₂O and Q⁺HCO₃^–·(x – 1)H₂O·HD are the lipophilic, deprotonated, and protonated ion-paired forms of pH indicator dye.^25,26 It follows from eq 7 that the absorbance of the film, due to Q⁺D^–·xH₂O, A, will be related to P_CO2 as described by eq 5, where α is the equilibrium constant for reaction 7 and R = [Q⁺HCO₃^–·(x – 1)H₂O·HD]/[Q⁺D^–·xH₂O].

One of the first reported CO₂-sensitive ink films comprised the deprotonated form of the pH indicating dye, meta-cresol purple, MCP, ion-paired with a tetraoctyl ammonium quaternary cation, from the lipophilic base, tetraoctyl ammonium hydroxide, TOAH, dissolved in a nonaqueous solvent in which were also dissolved the lipophilic polymer, EC, and a plasticizer, tributyl phosphate, TBP. An abbreviated formulation of this ink film is, therefore, MCP/EC/TOAH/TBP, i.e., dye, polymer, PTA, and plasticizer, and is used in the caption in Figure 1, along with other important test condition details. In this and all other inks described herein, the polymer (EC) was used to provide a lipophilic medium to dissolve the ion-pairs formed between the dye (MCP) and the PTA (TOAH) when the ink had been cast as a film and the solvent had evaporated. The plasticizer (TBP) of the polymer (EC) was used to improve the rate of permeation of CO₂ through the dried ink film.^28,29 The resulting MCP CO₂ indicator ink film was purple or yellow in the absence or presence of CO₂, respectively. A brief experimental detailing how the MCP ink film was made is given in section S2.1 of the SI file.

Figure 1.

(a) UV/vis absorption spectra of a MCP ink film recorded when exposed to different CO₂/N₂ gas mixtures, with the % CO₂ (from top to bottom) being 0.04, 0.3, 0.4, 1.0, 5.0, 10 and 100%, respectively, with the peaks at 605 and 420 nm, decreasing and increasing, with increasing % CO₂, respectively; (b) plot of R vs % CO₂ for the MCP indicator, where R was calculated using eq 5 and values of A taken from (a); (c) A vs time plots for the MCP indicator ink film on exposure to an alternating gas supply of air, ↑, and 5% CO₂, ↓. [MCP/EC/TOAH/TBP film; T = 20 °C; gas flow rate, f, 100 cm³/min; relative humidity, RH, 0%.] Adapted from ref (26). Copyright 1992 American Chemical Society.

Figure 1a illustrates the UV/vis absorption spectral changes exhibited by the MCP ink film when exposed to different % CO₂ levels. From the data in Figure 1a, the absorbance of this film, A, was determined as a function of % CO₂, which, via eq 5, was then used to generate the plot of R vs % CO₂ illustrated in Figure 1b. The straight-line nature of this plot is as predicted by eq 5, from the gradient of which values of α = 0.78% CO₂^–1 and P_CO2(S = 1/2) value (= 1/α) of 1.3% CO₂, were calculated.

Another key characteristic of a CO₂ indicator is its 90% response and recovery time, t_90↓ and t_90↑, respectively. In order to measure the latter, the absorbance of the MCP ink film was monitored as a function of time upon exposure to an alternating atmosphere of 0.04% (i.e., air) and 5% CO₂, the results of which are illustrated in Figure 1c and reveal t_90↓ and t_90↑ times of 2.6 and 31 s, respectively;²⁶ further practical details are given in S3 of the SI. Although the above t_90↓ and t_90↑ values are not particularly large, they prevent the indicator from being used for certain applications, such as in capnography, which requires response/recovery times of ca. ≪1 s.³⁰

The response and recovery times of this, and most CO₂ indicator films, depend simply upon the rate of diffusion of the CO₂ into and out of the indicator film, and not the kinetics of reaction 7.³¹ Consequently, they can be markedly shortened by making the films thinner or increasing the operating temperature.

3.1. Dependence of Sensitivity (α) on Dye pK_a and [Base]

As noted earlier, the sensitivity of the CO₂ ink film indicator depends directly upon the value of α, i.e., the gradient of the R vs % CO₂ plot, where, according to eq 4, α is inversely proportional to K_a and [Base]. Thus, early on in the development of CO₂-sensitive ink films, the variation in P_CO2(S = 1/2), = 1/α, was studied as a function of both K_a and [Base].³² In the former case, a number of different dyes, with different pK_a values, were used, namely, rosolic acid (RA), phenol red (PR), cresol red (CR), MCP, thymol blue (TB), ortho-cresol phthalein (OCP), and phenolphthalein (PP). A list of these dyes, their abbreviated names, pK_a values, and colors in their Q⁺D^–·xH₂O and Q⁺HCO₃^–·(x – 1)H₂O forms are given in section S4, Table S2 in the SI. The different ink films were formulated as described for the MCP ink film in section S2.1 in the SI, with the only altered parameter being the pH dye used. For each film, the associated value of P_CO2(S = 1/2) was determined from a R vs % CO₂ plot using the same method as described above for the MCP ink film and illustrated by the results in Figure 1a and b.

The plot of the results of this work, in the form of log{P_CO2(S = 1/2)} vs pK_a, is illustrated in Figure 2 and shows that, in accord with eq 4, the sensitivity (α) increases, and so P_CO2(S = 1/2) decreases, with a decreasing value of K_a (increasing value of pK_a). Equation 4 also predicts that a plot of log{P_CO2(S = 1/2)} vs pK_a should be a straight line with a gradient, m, of −1, which is represented by the broken line in Figure 2, from which it appears that this relationship only holds over the pK_a range 7–9, with dyes OCP and PP, and their >9 pK_a values, appearing as notable exceptions. The same deviation from linearity is also seen for the same dyes in aqueous solution³² and is due to the breakdown of the underlying assumption in eq 4, that the deprotonation of the bicarbonate to carbonate is not significant, as its pK_a is ca. 10.3. This assumption is not valid at very low %CO₂ values, and so m tends to −2, which helps explain the apparent steep departure from the broken line (m = −1) in Figure 2 exhibited by OCP and PP.³²

Figure 2.

Plot of log{P_CO2(S = 1/2)} vs pK_a, where the values of {P_CO2(S = 1/2)} were determined from R vs % CO₂ plots of the absorbance data derived from CO₂ indicator ink films employing the following, different pH indicator dyes, rosolic acid (RA), phenol red (PR), cresol red (CR), MCP, thymol blue (TB), ortho-cresol phthalein (OCP), and phenolphthalein (PP). [dye/EC/TOAH/TBP films; T = 20 °C; f, = 100 cm³/min, RH = 0%.] Adapted with permission from ref (32). Copyright 1994 Elsevier.

In other work, the dependence of α on 1/[Base], also predicted by eq 4, has been shown to hold for CO₂ indicator ink films.^32,33

3.2. Dependence of Sensitivity upon Temperature

All CO₂ colorimetric indicators, i.e., water-based, indicator ink, smart pigment, and plastic films, exhibit a decreasing sensitivity, α, with increasing temperature, implying an overall exothermic process, most likely linked to the dissolution of CO₂ in the encapsulation medium, which is known to be exothermic in both aqueous and nonaqueous solutions.³⁴Figure 3a provides a suitable illustration of this dependence for a PR ink film, in the form of a plot of the measured absorbance of the film, due to Q⁺PR^–·xH₂O as a function of % CO₂, recorded at different temperatures.³⁰ The data associated with each A vs % CO₂ profile in Figure 3a were used to generate a R vs % CO₂ plot, using eq 5, from which a value of α was derived. This data was then used to construct the plot of the ln(α) vs 1/T illustrated in Figure 3b, which fitted the basic thermodynamic expression,

8

where the product, 100α, is the equilibrium constant for reaction 5 in units of atm^–1. From the plot in Figure 3b and eq 8, values of ca. −88 kJ mol^–1 and −266 J mol^–1 K^–1 were calculated for ΔH and ΔS, respectively, for reaction 5 values which are consistent with those reported for other ink films²⁶ and with reaction 7 and its expected exothermicity.

Figure 3.

(a) Relative absorbance, A_rel (at λ(max) for D^–), as a function of % CO₂ for a PR ink film, recorded at the following temperatures (from bottom to top): 17, 28, 37, 46, and 55 °C, respectively; (b) plot of the values of ln(α), derived from the data and eq 5, in (a), vs 1/T. [PR/EC/TOAH/TBP films; T = 20 °C; f = 100 cm³/min, RH = 0%.] Adapted with permission from ref (30). Copyright 1997 Elsevier.

It follows from the above discussion that a CO₂ indicator film could be used as a temperature indicator, if the value of P_CO2 was set at some fixed value, P_CO2(fxd), and details of such a study are given in section S5 of the SI and ref (35).

3.3. Applications

The above work shows clearly that CO₂ ink films can be used to provide quantitative information regarding the level of CO₂ in the ambient gas phase. Subsequent studies were then carried out to explore their potential areas of application, the details of which are given below.

3.3.1. Capnography

In section 3, it was noted that, for a MCP ink film, although only a few seconds, the t_90↓ and t_90↑ values were still too big for monitoring CO₂ accurately in breath, i.e., capnography, which requires response/recovery times of ca. ≪1 s.³⁰ However, subsequent work on a PR ink film showed that both t_90↓ and t_90↑ decreased markedly with increasing temperature (see Figure 4a) and this feature allowed the indicator, when operated at 50 °C, to produce % CO₂ versus time profiles for real breath cycles that were near-identical to those recorded by a commercial capnometer, as illustrated in Figure 4b and c, respectively.³⁰

Figure 4.

(a) Plot of the measured t₉₀ values for response (solid line) and recovery (broken line) times for a PR ink film when exposed to an alternating atmosphere of 0 and 5% CO₂, as a function of indicator film temperature; % CO₂ vs time real breath profiles recorded using (b) the PR ink film and (c) a commercial capnometer. [PR/EC/TOAH/TBP films; T = 20 °C; f = 100 cm³/min, RH = 0%.] Adapted with permission from ref (30). Copyright 1997 Elsevier.

3.3.2. Quality Control of Carbonated Drinks

One potential area of application of a CO₂ indicator is in the measurement of the high P_CO2 in a carbonated drink, i.e., as a fizziness quality control indicator. The production of carbonated beverages is a billion-dollar industry in which it is essential that every bottle contains the appropriate high level of CO₂, typically ca. 4 bar. However, at present, there is no simple, inexpensive method to measure the P_CO2 in carbonated drink bottles.

A fizziness indicator would benefit not only the packager but also the retailer and consumer, as it would flag leaky bottles on the supermarket shelf and inform the consumer if a previously opened bottle has lost its sparkle, i.e., is flat. Thus, a fizziness indicator was developed based on a water-based ink with PR, sodium hydroxide, poly(vinyl alcohol), PVA, and glycerol, as the pH indicating dye, base, polymer, and plasticizer, respectively.³⁶ A water-based ink was selected because, when using the same dye, such inks are significantly less sensitive than their solvent ink counterparts, most likely due to the much lower solubility of CO₂ in water (and the hydrophilic polymer PVA) than in organic solvents (and hydrophobic polymers, such as EC).

For example, a typical solvent-based PR ink film has a P_CO2(S = 1/2) value of 7.4% (see Figure 2), whereas for a water-based PR ink film it is ca. 150%, i.e., ca. 1.54 bar!³⁶ The colors and spectra of this fizziness indicator, when exposed to different high P_CO2 values, are illustrated in Figure 5a and b. Figure 5c shows the fizziness indicator in a carbonated drink bottle when the liquid was either fully carbonated or flat.³⁶ Although of promise, since the ink used in this work is water-based, dye leaching is a problem, and so, in practice, in a carbonated drink bottle, it would need to be laminated with a GPM to prevent dye leaching or replaced with a plastic film equivalent, vide infra.

Figure 5.

(a) Photographs of a PR/NaOH/PVA/glycerol water-based CO₂ indicator film and (b) spectra (from top to bottom) as a function of increasing P_CO2; (c) photographs of (from left to right) a fizzy and flat carbonated drink. [PR/PVA/NaOH/films; T = 20 °C; f = 0 cm³/min, RH = 0%.] Adapted with permission from ref (36). Copyright 2011 Royal Society of Chemistry.

3.3.3. Manometry

An ink film that is sufficiently sensitive to respond to 0.04% CO₂ can function as an air pressure, P_air, sensor, provided the temperature is fixed since the value of P_CO2 is directly proportional to P_air. One area of possible application is in vacuum packaging (VP), which is commonly used in wholesale and retail food packaging, since there is no simple, inexpensive method for measuring the vacuum pressure inside such packages.

A vacuum pressure indicator, based on a CO₂-sensitive OCP ink, has been reported,³⁷ and Figure 6a illustrates the observed variation in color exhibited by the OCP ink film as a function of air pressure over the range 0–1 atm at 22 °C. From the measured absorbance, A, of the film recorded at different P_air,, the plot of R vs P_air, illustrated in Figure 6b, was generated, from which a value of P_air at which the OCP indicator film is halfway through its color change, i.e., P(S = 1/2), of ca. 0.62 atm was derived. Given the vacuum pressure in a typical VP food product is ca. 0.04 atm,³⁸ the above P_air(S = 1/2) value of ca. 0.62 atm determined for the OCP indicator shows it is well-suited for monitoring the pressure inside VP products.

Figure 6.

(a) Photographs of an OCP CO₂ indicator at different vacuum air pressures, P_air, at 22 °C, and (b) subsequent plot of the associated absorbance data in the form of R, calculated using eq 5, vs P_air, with a line of best fit of gradient 16.1 atm^–1. [OCP/EC/TBAH/TBP films; T = 22 °C; f = 0 cm³/min, RH = 0%.] Adapted with permission from ref (37). Copyright 2019 Royal Society of Chemistry.

The OCP CO₂ indicator ink film is appropriate for monitoring air pressure values well below 1 atm but not for monitoring levels at or above 1 atm. Thus, in a subsequent study a general-purpose air pressure indicator was made based on a TB ink film that was capable of measuring P_air values from 0.1 to 14 atm.³⁹

3.3.4. Dissolved CO₂

The lipophilic, solvent-based CO₂ indicator inks described in this section have also been used to measure the level of CO₂ dissolved in water, although they tend to quickly (within 1 h) change permanently to the color of the protonated form of the dye, HD, when placed in acidic solution, or in ones that contain a high concentration (1 M) of a salt, such as NaCl. This change in color, and loss in function, is due to the anion, A^–, in the acidic or salty solutions, exchanging with D^– in the ion-pair, Q⁺D^–·xH₂O in the film, to form Q⁺A^–·xH₂O, forcing D^– to form the lipophilic species HD;⁴⁰ a transition in form and color that is effectively permanent as Q⁺A^–·xH₂O is very stable. The situation is improved by replacing the usual TOAH base with tetradodecyl ammonium hydroxide or tetrakisdecylammonium hydroxide;⁴¹ but for long-term use, CO₂ indicator ink needs to be covered with a GPM.

4. CO₂ Indicator Pigments and Plastic Films

Although a number of solvent-based CO₂ indicator inks have been developed, they have not had much impact commercially, because of the printing ink industry change from solvent- to water-based inks. The commercial viability of CO₂ indicators has been improved with the development of the CO₂-sensitive plastic films, which are waterproof and can be produced cheaply using a scalable process. A typical CO₂-sensitive plastic film comprises a CO₂-sensitive pigment embedded in an inert, low melting point polymer film, such as low-density polyethylene, LDPE, by extruding a mixture of the pigment and polymer together. The pigment comprises nanoparticulate silica powder particles coated with a mixture of the dye and PTA,⁴²⁻⁴⁴ and so its formulation can be abbreviated to, dye/PTA/SiO₂, and that of the final, extruded plastic film as dye/PTA/SiO₂-LDPE. The preparation and characterization details for a MCP/TBAH/SiO₂ pigment and MCP/TBAH/SiO₂-LDPE film, along with those of an MCP/TBAH/EC/TBP ink film, are given in sections S2 and S6 of the SI. All three exhibit a similar sensitivity and so same, α, values, although the plastic film is slower in response and recovery. Reasons for the latter and further discussion of the differences between the ink and plastic film indicators are given in S6 in the SI. Note that the response of a plastic film indicator depends upon dye pK_a, [Base], and temperature in the same way as its ink film counterpart, and as described in sections 3.1 and 3.2, respectively. Thus, to minimize repetition, only potential applications of plastic films are considered below.

4.1. Applications

CO₂-sensitive, plastic film indicators can be used to replace the applications identified in section 3.3 for their ink film counterparts. However, one of their most striking features is its ability to function in very wet, possibly highly saline, environments, such as found in seawater, food packaging (of meats, say), and wounds. Thus, examples of their use in such fields are given below.

4.1.1. Plastic Films for Measuring Dissolved CO₂

As noted earlier, bare (no GPM) CO₂ inks cannot be used to measure % CO₂ in a highly saline aqueous solution. In contrast, plastic film indicators are stable in even high ionic solutions, such as seawater, and this feature was illustrated using a TB plastic film indicator.⁴³ Photographs of the TB/TBAH/SiO₂ pigment and the final, extruded TB plastic film, TB/TBAH/SiO₂-LDPE, before (blue) and after (yellow) exposure to 100% CO₂ are illustrated in Figure 7a and b, respectively.⁴³ The absorbance, A, of the CO₂-sensitive TB plastic film was measured as a function of % CO₂, both in the gas phase and in a 0.6 M NaCl aqueous solution, and the A vs % CO₂ data were then used to construct the R vs % CO₂ plots illustrated in Figure 7c. From the latter plots, it is clear that the sensitivity of the TB plastic film is reduced significantly (by a factor of ca. 2.8) when used to measure % CO₂ in saline solution compared to that in a gas, with P_CO2(S = 1/2) = 0.18% and 0.063% in saline solution and the gas phase, respectively.⁴³

Figure 7.

Photographs of (a) the silica pigment, coated with TB/TBAH, and (b) the final, extruded CO₂-sensitive plastic film indicator, before (blue) and after (yellow) exposure to 100% CO₂; (c) plots of R vs % CO₂ generated using the TB plastic film in the gas phase (solid data points and line) or in a 0.6 M NaCl aqueous solution (open circle data points and broken line); the gradients of these two lines are 15.9 and 5.6%^–1, respectively. [TB/TBAH/SiO₂-LDPE, T = 20 °C; f = 100 cm³/min.] Adapted with permission from ref (43). Copyright 2016 Elsevier.

Other work shows that the loss of sensitivity that accompanies using a plastic film to measure % CO₂ in water rather than the gas phase is a common feature for all plastic film indicators and appears to be due to the reversible uptake of water by the plastic indicator films. However, most importantly, unlike its ink counterpart, the TB CO₂-sensitive plastic film indicator is indefinitely stable in salty aqueous solutions.⁴³ This initial (but reversible) loss of sensitivity exhibited by plastic films when used for dissolved CO₂ measurements is due to an associated drop in the solubility of CO₂ in the encapsulation medium, which decreases with increasing polarity and hydrogen-bonding character of the solvent/encapsulation medium. Indeed, in aqueous solution, the sensitivity of the TB CO₂ plastic film indicator is very similar to that of a water-based ink based on the same dye, TB/TBAH/PVA, but operating in the gas phase.

4.1.2. After Opening Freshness (AOF) Indicator

Insignia Technologies Ltd. have recently developed an “After Opening Freshness”, AOF, label, based on the CO₂-sensitive plastic film technology developed by the Mills group,⁴⁴ that is able to inform the consumer whether the food in an opened food package in the fridge is still fresh, and so safe to eat.⁴⁵ Photographs showing the label and how it is incorporated into a typical refrigerator food package are illustrated in Figure 8a and b, respectively. Research shows that the AOF label illustrated in Figure 8 comprises a CO₂-sensitive CR plastic film sandwiched between two polymer layers, one of which is a barrier layer that controls the rate of permeation of CO₂ from inside the CR film to outside the label. A detailed schematic illustration of the structure of the label is illustrated in section S7, Figure S5 of the SI.

Figure 8.

Photographs of (a) the AOF label and (b) the label in the refrigerator food package of ham; (c) plot of apparent absorbance, A’, vs time after opening for an AOF label. [CR/TBAH/SiO₂-LDPE, T = 5 °C; f = 0 cm³/min.] Adapted with permission from ref (46). Copyright 2018 Elsevier.

The AOF indicator is effective when used with modified atmosphere food packages, MAP, since the high level of CO₂, typically >20%, usually used in MAP, is sufficient to change the color of the CR indicator film in the AOF label from purple, its color in air, to beige. In a MAP sealed package of ham, with % CO₂ = 25%, the CR indicator in the AOF label is initially beige but changes color when the package is opened, since the ambient level of CO₂ then drops to 0.04%. Under these circumstances, in the absence of the barrier film covering, the CR plastic film would normally regain its original (purple) color within ca. 26 min; but when incorporated in the AOF label, a diffusion-barrier film cover layer of 30 μm polyethylene terephthalate, PET, slows the rate of CO₂ permeation from inside to outside the label to such an extent that, at 5 °C, the typical temperature of a household fridge, the CR plastic film in the label takes about 4 days to turn purple. As illustrated in Figure 8a, in the AOF indicator, the color change of the central CR indicator “dot”, from beige to purple, is taken as an indication that the food in the opened package is “past best”.⁴⁶ Other work shows that it is possible to analyze the color of a photograph of the CO₂ indicator to derive a value for its apparent absorbance, A’, that is directly proportional to its real absorbance, A; this process is referred to here as digital camera colorimetry, DCC.⁴⁷ Further details of DCC and its use with CO₂ indicators are given in section S8 in the SI. Thus, Figure 8c shows how A’, for the AOF label, varies as a function of after opening time and highlights the time periods when the label is in its 3 different colored forms, and so signaling, “fresh” (beige), “still fresh” (brown), and “past best” (purple).

4.1.3. Early Wound Infection Indicator

Chronic wounds require careful tending and continuous monitoring, and in the UK alone, the annual cost of managing chronic wounds is over £3.2 billion. Unfortunately, by the time signs of infection appear in a chronic wound, the infection has often taken hold. Interestingly, there is strong evidence that infection is associated with surpassing a critical colony threshold, CCT, of ca. 10⁶ colony forming units per gram of tissue, i.e., CFU g^–1, regardless of the infecting species. Unfortunately, for reasons of cost, the measurement of microbial load is not a routine part of wound monitoring.

Currently, most current chronic wound dressings are occlusive, i.e. sealed, so as to reduce the chance of infection from outside,⁴⁸ but, since such a dressing seals in the headspace above a wound, it follows that a CO₂-indicator could be used to identify any appreciable increase in CO₂, above the ambient value of ca. 0.04%, which might be expected if the wound were infected.

Recently, a noninvasive, inexpensive, easy to use early wound-infection indicator, based on a 3D printed, CO₂ plastic film, has been reported which signals when the microbial load associated with a wound approaches 10⁶ CFU g^–1.⁴⁸ The indicator used a xylenol blue, XB, plastic film and was tested using an infected wound model based on pig skin, a schematic of which is illustrated in Figure 9a. The color of the XB plastic film indicator was then monitored, via photography, as a function of t, for different initial inoculum loadings, and the results of a typical set of runs are illustrated in Figure 9b, which show, not surprisingly, that the higher the initial level of the inoculum, the faster the production of CO₂ in the headspace. The photographs associated with runs C1 and C2 illustrated in Figure 9b also show that the indicator did NOT change when the wound bed was not inoculated (run C1) or if undamaged (no lacerations) pigskin was inoculated (run C2).⁴⁸

Figure 9.

(a) Schematic illustration of the set up used to test a XB plastic film as an early wound infection indicator, comprising (i) clear wound dressing plastic adhesive film, (ii) XB plastic film CO₂ indicator, (iii) gauze, (iv) infected pig-skin, (v) wet absorbent pad, and (vi) inert support substrate (50 μ PET); (b) photographs of the XB film in (a), and (c) measured variations in A’, determined from the images in (b), as a function of incubation time, t, for the pigskin inoculated with different loadings of Pseudomonas aeruginosa (CFU mL^–1). [XB/TBAH/SiO₂-LDPE, T = 30 °C; f = 0 cm³/min.] Adapted with permission from ref (48). Copyright 2022 Royal Society of Chemistry.

From each of the A’ vs t profiles illustrated in Figure 9c, the time taken for A’ to fall to a value of 0.6 was determined, t(A’ = 0.6). In a separate but otherwise identical set of experiments to those used to generate the data in Figure 9, for each inoculum, the value of the total microbial load on the pigskin, units CFU g^–1, was measured as a function of t, from which it was possible to determine the time taken for the load to reach the CCT value of 10⁶ CFU g^–1, t(10⁶). The subsequent plot of t(A’ = 0.6) vs t(10⁶) yielded a good straight line with a near unity gradient, suggesting that the XB plastic film can be used to signal when the bacterial load on the wound is at or near ca. 10⁶ CFU g^–1.⁴⁸ Other work showed that the XB indicator worked equally well with other anaerobes associated with wound infection, such as Enterococcus. faecium, Acinetobacter baumannii, Streptococcus pyogenes, Candida albicans, and Staphylococcus aureus.

4.1.4. A CO₂ Indicator Adhesive

Recently, it has been demonstrated that a CO₂-sensitive pigment can be easily incorporated into a pressure sensitive adhesive, PSA.⁴⁹ Such a smart adhesive is simple to make and can be readily applied to a plastic film to make very inexpensive CO₂-sensitive labels or tape.

5. Outlook

A brief comparison between the four different CO₂ indicator types reported here, namely, aqueous, ink, plastic, and adhesive films, is given in section S9, Table S4 in the SI. From this table, the outlook of the CO₂ sensitive plastic film and adhesive indicators appears particularly promising, especially when coupled with DCC,⁴⁷ which can be carried out with just a mobile phone and app. Although most work on CO₂ indicators is focused on their use in food packaging, this Account shows they have potential to be used in many different areas, such as in capnography, thermometry, manometry, wound monitoring, and environmental monitoring.

Biographies

Andrew Mills is a professor in the School of Chemistry and Chemical Engineering at Queens University Belfast. His research interests include semiconductor and dye photochemistry and optical sensors.

Lauren McDonnell received her MSci in Chemistry from Queens University Belfast (2021). She is a PhD student with research interests in 3D printed optical indicators, particularly those for the detection of CO₂.

Dilidaer Yusufu has a BS degree from University of Science and Technology of China (2013) and a PhD on novel optical CO₂ sensors from Queen’s University Belfast (2019). Her research interests focus mainly on the development of colorimetric and fluorescence CO₂ sensors.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/accountsmr.2c00226.

• Table of previous landmark papers, ink, pigment and plastic film indicator formulations, details on measuring response/recovery times, table of dyes used, details of a CO₂ indicator based temperature probe, comparison of pigment, ink and plastic film indicators, AOF indicator schematic, outline of DCC applied to CO₂ indicators and table comparing the characteristics of all 4 indicator types (PDF)

Author Contributions

The manuscript was written through contributions of all authors. All authors have approved the final version of the manuscript.

This work was funded by EPSRC, Grant ref No. EP/V041541/1.

The authors declare no competing financial interest.