Anesth Analg 2005;101:748-752
© 2005 International Anesthesia Research Society
doi: 10.1213/01.ANE.0000160530.77996.F2
TECHNOLOGY, COMPUTING, AND SIMULATION
Barium Hydroxide Lime Turns Yellow After Desiccation
Christofer D. Barth, MD*,
Marshall B. Dunning, III, PhD
,
Lynn Bretscher, PhD
, and
Harvey J. Woehlck, MD
*Department of Anesthesiology, Cleveland Clinic, Cleveland, Ohio; and Division of Pulmonary/Critical Care Medicine and Departments of Biochemistry and Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin
Address correspondence and reprint requests to Marshall B. Dunning III, PhD, Division of Pulmonary/Critical Care Medicine, Medical College of Wisconsin, 9200 W. Wisconsin Ave, Milwaukee, WI 53226. Address e-mail to mdunning{at}mcw.edu.
 |
Abstract
|
|---|
Ethyl violet is added to carbon dioxide absorbents and normally serves as an indicator of absorbent exhaustion. During the course of several prior studies of anesthetic breakdown, we noted (but did not publish) that barium hydroxide lime (BL), but not soda lime, turns yellow upon desiccation. We hypothesize that ethyl violet undergoes chemical reaction to produce a yellow colorant in desiccated BL. We qualitatively studied the time course of yellow color development during desiccation of these absorbents with dry oxygen. The yellow colorant was extracted from desiccated absorbent with diethyl ether, separated with chromatography, and analyzed with proton nuclear magnetic resonance and combined gas chromatography and mass spectrometry. The yellow color develops after BL has reached nearly complete desiccation. We successfully identified that ethyl violet decomposes into the yellow colorant 4,4'-bis(diethylamino)benzophenone upon desiccation of BL. The color is not intense, is not useful for identifying low levels of absorbent desiccation, and may be difficult to see through tinted canisters. It may be possible for BL to be sufficiently desiccated to allow chemical breakdown of anesthetics, but not yet show yellow coloration. However, if yellow coloration exists, one should assume that it has become desiccated.
|
Introduction
|
|---|
Ethyl violet [CAS 2390-59-2] is a pH-sensitive triarylmethane dye. It is added to most carbon dioxide (CO2) absorbents sold in the United States and normally serves as an indicator of absorbent exhaustion by CO2. Ethyl violet in fresh barium hydroxide lime (BL) and soda lime (SL) has a hydroxyl group attached to the central carbon and is colorless. CO2 reacts with and consumes the hydroxides, resulting in a lower absorbent pH, removal of the hydroxyl group from the central carbon of ethyl violet, and a color change to blue/violet.
Ethyl violet may fail to indicate that the absorbent material is exhausted. In clinical situations, sufficient light exposure may result in photodeactivation (1), and the chemical literature notes similar photo-reactions with substituted triarylmethane dyes (2). Desiccation of SL and Amsorb, a CO2 absorbent without alkali metal hydroxides, may also result in blue/violet coloration (3).
BL and SL are approximately 15%–20% water by mass. In anesthetic machines, prolonged fresh gas flow can desiccate absorbents. Desflurane, isoflurane, enflurane, halothane, and sevoflurane react with these desiccated absorbents to produce carbon monoxide (4); life-threatening carbon monoxide poisonings have been reported with desflurane (5). Sevoflurane may also react with desiccated absorbents to generate heat, formaldehyde, methanol, and, under more extreme conditions, explosive combustion (6–8). Therefore, the detection of desiccated absorbents may be clinically useful and enhance patient safety.
There are no accepted standards for detecting BL and SL desiccation. We hypothesized that ethyl violet undergoes a chemical reaction to produce a yellow colorant in desiccated BL, and we investigated whether this color change could be a useful marker for detecting desiccation.
|
Methods
|
|---|
BL (Baralyme® CO2 Absorbent Granules; Chemetron Medical Division, Allied Healthcare Products, Inc., St. Louis, MO) and SL (Sodalime Carbon Dioxide Absorbent Granules; Puritan Bennett®, Pleasanton, CA) were desiccated in canisters taken from a Draeger anesthesia machine. Desiccation was provided by the upward flow of dry oxygen at 10 L/min for 60 days delivered by a freestanding flowmeter outside of the anesthesia machine. Absorbent samples of 1200.0 ± 0.1 g (8 and 4 samples of BL and SL, respectively) were weighed at zero, approximately 5, approximately 10, 30, and 60 days. Color cards (Royal Horticultural Society Color Chart; Royal Horticultural Society, London, UK) were compared side-by-side to the absorbent under identical lighting conditions by two non-color blind individuals at the indicated time intervals, and recorded. Colors are also reported as red, green, blue (RGB) values that were subsequently calculated from the color cards by Medical Graphics Inc. (Milwaukee, WI.) Three undesiccated (control) samples were treated similarly in terms of light and temperature exposure, without the dry oxygen flow. Samples and controls were exposed to indirect sunlight and daytime ambient fluorescent overhead room lights approximately 12 h per day throughout the desiccation and measurement period. At the termination of the experiment, the color in yellow absorbent was tested for water fastness by pouring approximately 240 mL of water over the desiccated yellow absorbent, and observing the color immediately and after 24 h. This quantity of water made the absorbent wet to touch, which ensured adequate distribution of the water throughout the absorbent granules.
Colorants were extracted from samples of BL and SL (desiccated for 60 days) with anhydrous diethyl ether. The ether was filtered, evaporated under vacuum, and the concentrated ether solution was loaded onto silica gel thin-layer chromatography plates, and eluted with ethyl acetate/hexanes 3:1. A yellow band (Rf 0.6) appeared. Further purification of this yellow band was done by thin-layer chromatography with an eluent of ethyl acetate/hexanes 1:1. The compound was analyzed using proton nuclear magnetic resonance (300 MHz; dissolved in deuterated chloroform, chloroform reference peak). Additional analysis was performed using gas chromatography-electrospray ionization-mass spectrometry (mobile phase was 9:1 acetonitrile/water containing 0.005% acetic acid, drying gas temperature was 60°C, fragmentor was 70 V, detection was made in a positive mode).
Undesiccated 50-g samples of BL and SL served as controls. Colorants were extracted with 100 mL of diethyl ether. These solutions were filtered, concentrated, loaded similarly onto silica gel thin-layer chromatography plates, and eluted with ethyl acetate/hexanes 1:1.
|
Results
|
|---|
BL and SL were significantly desiccated after 5 days of drying. Only a small additional decrease in mass was detected at 60 days. BL lost less mass than the SL.
Color change occurred gradually over 60 days of desiccation. BL samples had a faint yellow color by 4 days that evolved to a maximal yellow color before 60 days (Table 1). SL developed a faint blue color after 10 days and had no significant yellow coloration at any time. Controls remained colorless throughout the 60 days. The yellow color of desiccated BL persisted after rehydration. Aqueous extracts of previously desiccated BL had a yellow color.
Thin-layer chromatography separated extracts from desiccated BL and SL into several fluorescent bands. In the extract from BL, the top band reacted with light, turning black; elution of extracts of this top band in ethyl acetate/hexanes 1:1 revealed there to be 2 distinct photoreactive substances. The middle band (Rf 0.45) appeared yellow in color, was easily repurified, and is further characterized below. The lowest band (Rf 0.36) appeared red-orange and proton nuclear magnetic resonance spectroscopy revealed no distinct aromatic peak but a broad peak at 
5.5–7.0 ppm and at least 2 peaks in the 1.0–2.0 ppm range. Extracts of desiccated SL separated into a top bluish band, which may have reacted with light; a middle yellow band (Rf 0.42); and a lower bluish band. Notably, the intensity of the middle yellow band was consistently greater for the desiccated BL compared with those from desiccated SL. Analysis of the yellow colorant revealed loss of an aryl group, and suggested presence of a hydroxyl group on the central carbon. The proton nuclear magnetic resonance and mass spectrometry results from the yellow compound isolated from SL and BL were identical to that of pure 4,4'-bis(diethylamino)-benzophenone [CAS 90-93-7] (Aldrich Chemical Co., Milwaukee, WI; catalog no. 160326-25G). This product appears to be relatively stable and unaffected by ambient light.
Untreated BL and SL did have a yellow band of low intensity at a similar Rf to the yellow compound isolated from desiccated absorbents. Pure ethyl violet (Aldrich Chemical Co.), however, did not produce a yellow band when treated under these conditions.
|
Discussion
|
|---|
This report documents that yellow coloration of BL indicates nearly complete desiccation. BL and SL are significantly desiccated by 10 L/min of oxygen flow in 48–66 hours (9), although our measurements began at 5 days when desiccation was nearly complete. The production of a yellow colorant occurred only with BL, began only at very small water content, and progressed slowly. The development of yellow color appeared to be unaffected by ambient light, although we did not control for this variable. This study, nonetheless, applies to typical lighting conditions in operating rooms.
A distinction should be made between desiccation, which is the loss of water, and exhaustion, which is the absorption of CO2. Exhaustion is the conversion of alkaline earth and alkali hydroxides to carbonates. Both desiccation and exhaustion may produce a blue or violet coloration of SL (3). The color change on exhaustion of hydrated absorbents is usually intense. No studies have been published regarding color changes in absorbents that are both exhausted and desiccated. Color change of SL on desiccation was a subtle violet, consistent with the findings of Knolle et al. (3). We noted that bluish coloration of SL was not intense even after 60 days of desiccation, and consequently is also a clinically poor marker of desiccation. Although small amounts of yellow colorant were produced by SL after desiccation, the very faint color may have been obscured by the larger amount of the more intense but contrasting blue/violet color, and hence was not noticeable to the human eye. The yellow coloration of desiccated BL is not intense at first, but intensifies with time after desiccation is complete. Because the color change takes time after desiccation has occurred, our results do not contradict the finding of Knolle et al. (3) who noted no color change of Baralyme® after only 6 minutes of dry heating. Knolle et al. also identified that the more sodium hydroxide an absorbent contained, the less its tendency to turn blue on desiccation, and that potassium hydroxide completely prevented blue coloration development on desiccation.
The isolated yellow compound was produced more readily in BL than SL. Both the yellow coloration of bulk BL and the intensity of the yellow band on thin-layer chromatography from BL extracts were consistently more than those resulting from SL. The techniques used to isolate and purify the yellow compound did not lead to its production because pure ethyl violet did not break down under these solvent conditions. The qualitatively small amount of the yellow compound in undesiccated absorbents suggests that ethyl violet may be slightly unstable in BL and SL, but this process seems to be significantly enhanced by desiccation. The yellow colorant identified matched the color of the yellow absorbent when viewed by the naked eye, and was concluded by the authors to be the responsible pigment. The black and red bands identified on chromatography were not further studied because they did not match the color of the yellow absorbent, and were present in very small quantities on chromatography.
We identified that ethyl violet decomposes into 4,4'-bis(diethylamino)benzophenone in desiccated absorbent. This reaction has also been described in solution phase (10), and the authors (unpublished data) also were able to produce yellow coloration in solution phase by reacting dilute solutions of ethyl violet with an excess of sodium oxide in anhydrous methanol in the dark. This suggests that anhydrous alkaline conditions are necessary for the breakdown of ethyl violet.
Based on the structures of ethyl violet and the yellow colorant, and attempting to incorporate the findings of Knolle et al. (3) and those from the present study, we propose the following explanation for our results. To break down ethyl violet, a stronger base is necessary than aqueous hydroxide, because ethyl violet is relatively stable in hydrated bases such as sodium hydroxide. Water exerts a solvent effect on strong acids or bases, termed "leveling," because all, including sodium or potassium hydroxides, are completely ionized in water. Because any stronger basic ion immediately reacts with water to form the hydroxide ion, the strongest base that can exist in water solution is the aqueous hydroxide ion (11). In nonaqueous systems such as methanol, alkoxides may become the predominant ion, and alkoxides are more basic than hydroxides based on their higher pKa. Oxide is also a much stronger base than hydroxide. Under anhydrous conditions, it becomes possible for catalytic amounts of oxide to be produced from hydroxide, as the absence of water drives Equation 2 (12) to the right.
|
|
If extremely strong bases such as alkoxide or oxide are present, or if the concentration of water is reduced to the point whereby water is no longer present in sufficient quantity to reprotonate carbanion or other weakly acidic intermediates, acid base reactions may occur between weakly acidic hydrogens and the basic ion. This has been proposed as the initial step in anesthetic-absorbent reactions (13) and we propose that this is also the initial step in ethyl violet breakdown to the yellow colorant (Figs. 1 and 2). Just as water inhibits anesthetic-absorbent reactions, this mechanism is consistent with the inhibition by water of yellow colorant formation, because the yellow colorant slowly develops only after complete desiccation has occurred. We did not identify the other proposed breakdown product of ethyl violet shown in Figure 2 (diethylaminobenzene); however, this compound is volatile and would have evaporated during the desiccation process. Our findings are also consistent with greater yellow colorant production by potassium hydroxide than sodium hydroxide, because, in the absence of water, heavier alkali metal hydroxides would have greater base strength than lighter alkali metal hydroxides (12), and would be predicted to generate greater reaction. In addition, potassium hydroxide forms a stable monohydrate in anhydrous medium that can provide the driving force for seemingly unfavorable reactions such as the deprotonation of weak acids with a pKa range of 16–27 (12). Although the blue coloration on desiccation of SL invokes only a dehydration/acid base reaction of ethyl violet from the leuko form to the violet form (3), the greater basicity of potassium hydroxide in anhydrous conditions compared with sodium hydroxide is consistent with the findings that potassium hydroxide keeps ethyl violet in the hydroxylated leuko form, preventing blue coloration on desiccation.
View larger version (15K):
[in this window]
[in a new window]
|
Figure 1. Proposed reaction mechanism for the destruction of the leuko phase of ethyl violet by strong base in anhydrous medium.
|
|
View larger version (9K):
[in this window]
[in a new window]
|
Figure 2. Proposed products for the breakdown of ethyl violet. The compound to the left is 4,4'-bis(diethylamino)benzophenone, a yellow colorant that was isolated in this study. To the right is diethylaminobenzene, a volatile compound that was neither sought nor identified in this study.
|
|
There are limitations to the clinical significance of our study. Yellow coloration of BL is a specific but insensitive marker of complete prior desiccation. The clinical utility is small because BL may be completely desiccated, but the yellow coloration may not be apparent for days. Because the color change is faint, and tinted absorbent canisters themselves may prevent observation of this color change, it would be necessary to observe BL absorbent directly from the top of an open canister, rather than through the canisters, to detect this change. In a clinical situation, the presence of yellow color in CO2 absorbent indicates previous desiccation and the absorbent should be replaced. If the color change of BL could be enhanced such that yellow coloration would appear with slight desiccation, it may become a clinically useful indicator of desiccated CO2 absorbent. To prevent a false sense of security that nonyellow absorbents are always safe, we emphasize that yellow absorbents have been previously desiccated, although absorbent may be partially desiccated without color change. Although BL is no longer available for clinical use in the United States, other formulations of CO2 absorbent containing potassium hydroxide may become available in the future, and yellow coloration may be possible in those formulations.
|
Acknowledgments
|
|---|
The authors thank Dr. Owen Griffith, PhD, Professor of Biochemistry, Medical College of Wisconsin, for the use of his laboratory and Dr. Kasem Nithipatikom, PhD, Associate Professor, Department of Pharmacology and Toxicology, Medical College of Wisconsin, for the mass spectrometry results.
|
Footnotes
|
|---|
This study was supported by departmental funds of the departments of Anesthesiology and Biochemistry (Medical College of Wisconsin).
Accepted for publication February 2, 2005.
|
References
|
|---|
- Andrews JJ, Johnston RV, Bee DE, Arens JF. Photodeactivation of ethyl violet: a potential hazard of sodasorb. Anesthesiology 1990;72:59–64.[Medline]
- Jarikov VV, Neckers DC. Photochemistry and photophysics of triarylmethane dye leuconitriles. J Org Chem 2001;61:659–71.
- Knolle E, Linert W, Gilly H. The color change in CO2 absorbents on drying: an in vitro study using moisture analysis. Anesth Analg 2003;97:151–5.[Abstract/Free Full Text]
- Fang ZX, Eger EI II, Laster MJ, et al. Carbon monoxide production from degradation of desflurane, enflurane, isoflurane, halothane and sevoflurane by soda lime and Baralyme®. Anesth Analg 1995;80:1187–93.[Abstract]
- Berry PD, Sessler DI, Larson MD. Severe carbon monoxide poisoning during desflurane anesthesia. Anesthesiology 1999;90:613–6.[ISI][Medline]
- Holak EJ, Mei DM, Dunning MB III, et al. Carbon monoxide production from sevoflurane breakdown: modeling of exposures under clinical conditions. Anesth Analg 2003;96:757–64.[Abstract/Free Full Text]
- Wu J, Previte JP, Adler E, et al. Spontaneous ignition, explosion, and fire with sevoflurane and barium hydroxide lime. Anesthesiology 2004;101:534–7.[Medline]
- Castro BA, Freedman LA, Craig WL, Lynch C III. Explosion within an anesthesia machine: baralyme, high fresh gas flows and sevoflurane concentration. Anesthesiology 2004;101:537–9.[Medline]
- Woehlck HJ, Dunning M III, Raza T, et al. Physical factors affecting the production of carbon monoxide from anesthetic breakdown. Anesthesiology 2001;94:453–6.[Medline]
- Baptista M, Indig G. Mechanism of photobleaching of ethyl violet non-covalently bound to bovine serum albumin. Chem Commun 1997;18:1791–2.
- Huheey JE. Inorganic chemistry. 3rd ed. New York: Harper & Row, 1983:336–9.
- Cotton FA, Wilkinson G, Murillo CA, Bochmann M. Advanced inorganic chemistry. 6th ed. New York: John Wiley & Sons, 1999:66, 98, 445.
- Baxter PJ, Garton K, Kharasch ED. Mechanistic aspects of carbon monoxide formation from volatile anesthetics. Anesthesiology 1998;89:929–41.[ISI][Medline]
This article has been cited by other articles:
|
|
|
|
 
G. Cheng and M. Greenberg
International variances in carbon dioxide absorbent colour indicators
Can J Anesth,
March 1, 2007;
54(3):
243 - 243.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
