or

Download PDF
0 downloads 0 Views 2MB Size Report
Comment
Dec 4, 2015 - the solvent, whether heated using microwave radiation or ..... are not likely to be possible using microwave dielectric heating, but it may be ...
Communication

Microwave-Assisted Superheating and/or Microwave-Specific Superboiling (Nucleation-Limited Boiling) of Liquids Occurs under Certain Conditions but is Mitigated by Stirring Anthony Ferrari, Jacob Hunt, Albert Stiegman * and Gregory B. Dudley * Received: 28 October 2015 ; Accepted: 20 November 2015 ; Published: 4 December 2015 Academic Editors: Marilena Radoiu, Jean-Jacques Vanden Eynde and Annie Mayence Department of Chemistry and Biochemistry, Florida State University, Tallahassee, FL 32306-4390, USA; [email protected] (A.F.); [email protected] (J.H.) * Correspondence: [email protected] (A.S.); [email protected] (G.B.D.); Tel.: +1-850-644-2333 (G.B.D.)

Abstract: Temporary superheating and sustained nucleation-limited “superboiling” of unstirred liquids above the normal atmospheric boiling point have been documented during microwave heating. These phenomena are reliably observed under prescribed conditions, although the duration (of superheating) and magnitude (of superheating and superboiling) vary according to system parameters such as volume of the liquid and the size and shape of the vessel. Both phenomena are mitigated by rapid stirring with an appropriate stir bar and/or with the addition of boiling chips, which provide nucleation sites to support the phase-change from liquid to gas. With proper experimental design and especially proper stirring, the measured temperature of typical organic reaction mixtures heated at reflux will be close to the normal boiling point temperature of the solvent, whether heated using microwave radiation or conventional convective heat transfer. These observations are important to take into consideration when comparing reaction rates under conventional and microwave heating. Keywords: superheating; microwave; reflux; temperature; boiling point

1. Introduction Accurate temperature measurements are critical to understanding thermochemical processes. While this can be easily done for conventional convective heating using thermometers and thermocouples, it is not so easily done under microwave heating due to interactions of the radiation field with standard thermometers. There are two common approaches to measuring temperature inside the microwave cavity. One is to use remote, infrared (IR) sensors that measure the black-body emission of the system and derive the temperature from that. The second is the use of fiber optic thermometers that are unaffected by the radiation field and can provide accurate temperature determination inside the cavity. Ultimately, all temperature measurements are standardized against reproducible, physical phenomena, such as the melting and boiling points of water (cf. Centigrade scale). Even when accurate bulk temperature is determined, however, one must recognize and consider that microwave radiation creates heat through mechanisms that are distinct from those of convective heating, which potentially can result in selective heating and inhomogeneous temperature distributions that cannot be detected by bulk temperature measurements. Inhomogeneous temperature distributions in selectively heated systems must be inferred from other physical properties of the system. Reliable determinations of bulk temperature

Molecules 2015, 20, 21672–21680; doi:10.3390/molecules201219793

www.mdpi.com/journal/molecules

Molecules 2015, 20, 21672–21680

inside the microwave cavity are critical to understanding the unique impacts of microwave heating on thermochemical processes. We reported [1]—and later quantified [2,3]—microwave-specific rate accelerations of organic reactions due to selective heating of polar solutes dissolved in nonpolar solvents. Many reports of Molecules 2015, 20, page–page such phenomena have been dismissed as artifacts of improper temperature measurements [4], and our series of papers considerable [5–7] and some controversy [8–11]. Much of We stimulated reported [1]—and later quantified interest [2,3]—microwave-specific rate accelerations of organic dueto to confusion selective heating of polar dissolvedbetween in nonpolar heat solvents. Many reports of the controversyreactions related over thesolutes distinction and temperature, and to the such phenomena have been dismissed as artifacts of improper temperature measurements [4], and challenges of accurately measuring temperature microwave-heated solutions. To the latter our series of papers stimulatedbulk considerable interest [5–7]of and some controversy [8–11]. Much of the related optic to confusion over the distinction between heat and temperature, and to the point, we use controversy internal fiber temperature probes, external infrared sensors, and/or thermal challenges of accurately measuring bulk temperature of microwave-heated solutions. To the latter imaging cameras to record system temperatures in our experiments, and finally, we conducted point, we use internal fiber optic temperature probes, external infrared sensors, and/or thermal imaging recordphysical system temperatures our experiments, and finally,(toluene, we conducted experiments incameras whichto the boiling inpoint of the solvent atexperiments atmospheric pressure) which the physical boiling point of the solvent (toluene, at atmospheric pressure) determined the determined thein bulk solution temperature (Scheme 1). bulk solution temperature (Scheme 1).

Scheme 1. Recapitulation of thermal Friedel–Crafts benzylation reactions conducted in refluxing

Scheme 1. Recapitulation thermal benzylation reactions in refluxing toluene (reprintedof from [2]). UnderFriedel–Crafts otherwise-identical conditions, the reaction in whichconducted reflux was sustained application of microwave radiation was faster. Both reaction mixtures were in visibly toluene (reprinted fromby[2]). Under otherwise-identical conditions, the reaction which reflux was homogeneous and vigorously stirred. sustained by application of microwave radiation was faster. Both reaction mixtures were visibly The specific experiments recounted in Scheme 1 unambiguously point to a difference between homogeneous and vigorously stirred. microwave and convective heating on this reaction, but they do not alone clarify the origin of the difference. Based on this and a series of related experiments [1,2], we attributed the difference to selective heating of the ionic solute in (BnOPB) in the1 microwave reactor, which perturbs The specific experiments recounted Scheme unambiguously point to athermal difference between equilibrium between the solute and the bulk solvent (toluene). Interactions between microwave microwave andradiation convective heating on this reaction, but they do not alone clarify the origin of the and the ionic solute convert microwave electromagnetic energy into thermal energy (heat) difference. Based on this and a series of related experiments [1,2], we attributed the within nanometer-sized solute domains, resulting in thermal reactivity of the solute that exceeds difference to what would be expected based on the observable temperature of the bulk solvent. A more detailed selective heating of the ionic solute (BnOPB) in the microwave reactor, which perturbs thermal explanation of the underlying theory can be found in our recent Perspective Article on microwaveequilibrium between the solute and the specific reaction rate enhancement [12].bulk solvent (toluene). Interactions between microwave An alternative potential explanation of the facts laid out in Scheme 1 is thatenergy microwave-specific radiation and the ionic solute convert microwave electromagnetic into thermal energy solvent “superheating” [8] elevated the temperature of the toluene solution significantly above its (heat) within nanometer-sized solute domains, resulting in thermal reactivity atmospheric boiling point. Superheating phenomena in stirred liquids under microwave heatingof hadthe solute that been regarded as negligiblebased [13], but on then the they were later reported and described by same authors exceeds what would be expected observable temperature ofthethe bulk solvent. A more as being potentially confounding “even (when) applying vigorous stirring” [14]. In light of conflicting detailed explanation of the underlying theory can be found in our recent Perspective Article on reports and confusion surrounding various potential forms of superheating in microwave experiments, microwave-specific reaction enhancement we have examinedrate the issue more closely. We[12]. immediately ruled out this alternative explanation of the observations recounted in Scheme 1 by directly measuring in situ the reflux temperature of our An alternative potential explanation of the facts laid out in Scheme 1 is that microwave-specific (stirred) reaction mixture (Figure 1, black line) [2,8]. Our investigations and observations will help solvent “superheating” [8]theelevated the temperature the toluene superheating. solution significantly above its bring clarity to confusions surrounding the various of forms of microwave As noted above, superheating phenomena were negligible in our experiments atmospheric boiling point. Superheating phenomena in stirred liquids involving under stirred microwave heating liquids (Figure 1, black line). What can also be seen in Figure 1 (green line), however, is that in the absence had been regarded aswenegligible [13], but later and described by the of stirring, were able to document two then related they thermal were events in rapid reported succession: superheating [15,16] and “superboiling” of our toluene solution.vigorous These events— same authors as being potentially(nucleation-limited confounding boiling “even[17]) (when) applying stirring” [14]. In temporary superheating and sustained superboiling—have seemingly been conflated in some of the light of conflicting reports and confusion surrounding various potential forms of superheating microwave chemistry literature, but we wish to make a clear distinction between the two. For example, in microwave experiments, we have examined the issue more closely. We immediately ruled out only the latter is microwave-specific.

this alternative explanation of the observations recounted in Scheme 1 by directly measuring in situ the reflux temperature of our (stirred) reaction mixture (Figure 1, black line) [2,8]. Our investigations and observations will help bring clarity to the confusions surrounding the various 2 forms of microwave superheating. As noted above, superheating phenomena were negligible in our experiments involving stirred liquids (Figure 1, black line). What can also be seen in Figure 1 (green line), however, is that in the absence of stirring, we were able to document two related thermal events in rapid succession: superheating [15,16] and “superboiling” (nucleation-limited boiling [17]) of our toluene solution. These events—temporary superheating and sustained superboiling—have seemingly been conflated in some of the microwave chemistry literature, but we wish to make a clear distinction between the two. For example, only the latter is microwave-specific. 21673

Molecules 2015, 20, 21672–21680 Molecules 2015, 20, page–page

Figure 1. Plots of temperature over time for refluxing solutions of BnOPB in toluene under microwave

Figure 1. Plots of temperature over time for refluxing solutions of BnOPB in toluene under microwave heating (reprinted from [9]). The top (green) line shows the measured bulk temperature profile in the heating (reprinted from [9]). The top (green) line shows the measured bulk temperature profile in the absence of stirring. Initial superheating (circled) is followed by sustained superboiling of the unstirred absence of stirring. Initial superheating (circled) is followed by sustained superboiling of the unstirred liquid. Superboiling was visibly chaotic, consistent with the erratic temperature profile. The lower two liquid. Superboiling was visibly chaotic, consistent the erratic either temperature profile. boiling The lower lines show the measured bulk temperature profiles ofwith stirred solutions, with or without two lines show the measured bulk temperature profiles of stirred solutions, either with or without chips. These refluxing solutions appeared qualitatively similar to refluxing solutions under conventional boiling chips. These refluxing solutions appeared qualitatively similar to refluxing solutions under heating to which they were compared. conventional heating to which they were compared. Superheating of liquids above the standard boiling point can be achieved without the use of microwave energy, although microwave heating facilitates the process. The temperature of a liquid Superheating of liquids above the standard boiling point can be achieved without the use of can be elevated to above its boiling point, provided that care is taken to avoid nucleation sites that microwave energy, although microwave heating facilitates the process. The temperature of a liquid could help trigger the liquid→gas phase change. The superheated liquid will exist in this metastable can be elevated above itsresulting boilinginpoint, provided that caregas is bubble(s). taken to avoid nucleation sites that state until it istoperturbed, the formation of the initial Once boiling is initiated, couldhowever, help trigger the liquidÑgas phasetochange. The superheated liquid(and willsometimes exist in this metastable the metastable liquid ceases exist, and excess heat is quickly violently) state liberated until it to is the perturbed, resulting in the formation of the the initial bubble(s). Once boiling is surroundings. In conventionally heated systems, bulkgas temperature of the remaining liquid rapidly regresses to the normal boiling point expected for the ambient atmospheric conditions. initiated, however, the metastable liquid ceases to exist, and excess heat is quickly (and sometimes boiling), in contrast to solvent superheating, has bulk been reported violently)Superboiling liberated to(nucleation-limited the surroundings. In conventionally heated systems, the temperature as remaining a microwave-specific phenomenon related to the inability of the bulkexpected solvent tofor undergo a of the liquid rapidly regresses to the normal boiling point the ambient liquid→gas phase change fast enough to remove the radiation-generated bulk heat. This novel atmospheric conditions. phenomenon has also been called “super-heated boiling” [18] or “superheating” [8,19], although the Superboiling (nucleation-limited boiling), in contrast to solvent superheating, has been reported latter term is perhaps best reserved for the more general (and not microwave-specific) phenomenon as a described microwave-specific phenomenon related to the inability of the bulk solvent to undergo a in the previous paragraph. We prefer the term “superboiling” for its simplicity and for how liquidÑgas phase change fast enough to remove the radiation-generated bulk heat. This novel it reflects the distinction from conventional superheating of static liquids. phenomenon has also been called “super-heated boiling” [18] or “superheating” [8,19], although the Like other reported microwave-specific phenomena, superboiling is perhaps neither recognized latternor term is perhaps bestwithin reserved fororganic the more general (and not much microwave-specific) phenomenon accepted broadly in the community. As with of the early literature on microwave the effects microwave heating, early reports mayfor be how described in thechemistry previous and paragraph. Weofprefer the term “superboiling” forofitssuperboiling simplicity and subject skepticism from owingconventional to conflicting superheating data and to theoffundamental challenges associated with it reflects thetodistinction static liquids. reproducibility and accurate temperature measurement. our opinion,isitperhaps can be difficult the Like other reported microwave-specific phenomena, In superboiling neither for recognized casual reader to discern which reported phenomena are likely to be reproducible and which are likely nor accepted broadly within in the organic community. As with much of the early literature on experimental artifacts. As noted above, we previously documented superheating and superboiling microwave chemistry and the effects of microwave heating, early reports of superboiling may be in unstirred toluene solutions, and we showed that vigorous stirring ameliorated both of these effects subject to skepticism to conflicting data andusing to the fundamental associated in toluene. Here weowing report the results of experiments common alcoholic challenges solvents. Taken together, with reproducibility and accurate temperature measurement. In our opinion, it can be difficult for the our observations qualitatively validate previous reports of superboiling while underscoring the casual reader to of discern which reportedexperimental phenomenadesign are likely to be and which are likely importance potentially overlooked details on reproducible reaction outcomes.

experimental artifacts. As noted above, we previously documented superheating and superboiling in unstirred toluene solutions, and we showed that vigorous stirring ameliorated both of these effects in toluene. Here we report the results of experiments using common alcoholic solvents. Taken 3 together, our observations qualitatively validate previous reports of superboiling while underscoring the importance of potentially overlooked experimental design details on reaction outcomes. 21674

Molecules 2015, 20, 21672–21680

Molecules 2015, 20, page–page

2. Experimental Designs 2. Experimental Designs All reflux experiments and measurements were conducted on a 30-mL volume of alcohol solvent All reflux experiments and measurements on a 30-mL of alcohol solvent in a 50-mL quartz round-bottom flask with a were 24/40conducted joint connected to a volume reflux condenser, except in a 50-mL quartz round-bottom flask with a 24/40 joint connected to a reflux condenser, except where where noted. Heating was accomplished using a CEM Discover (R) SP2 2.45 GHz microwave system noted. Heating using a CEM Discover (R)where SP2 2.45 GHzon microwave system operating operating at 75 was W ofaccomplished applied power. Liquids were stirred, noted, the highest stirring setting at 75 W ofinapplied power. Liquids wereReagent-grade stirred, where noted, on theethanol, highest stirring setting possible the possible the CEM reactor system. methanol, and isopropyl alcohol in were CEM reactor system. Reagent-grade methanol, ethanol, and isopropyl alcohol were used as received. used as received. thethe liquids werewere measured internally with a Neoptix fiber optic The temperatures temperaturesofof liquids measured internally with a (R) Neoptix (R) temperature fiber optic probe that was interfaced withinterfaced the microwave, except where noted. The fiber noted. optic thermometer was temperature probe that was with the microwave, except where The fiber optic ˝ calibrated against a NIST traceable thermocouple accurate to ±0.001 °C. thermometer was calibrated against a NIST traceable thermocouple accurate to ˘0.001 C. 3. Results Resultsand andDiscussion Discussion The first set of heating experiments featured unstirred liquids. As shown in Figure 2, methanol, ethanol, and isopropyl alcohol were each heated using 75 W of applied microwave power for 3 min. In all cases, temperatures significantly above the normal atmospheric boiling solvent were recorded using the in situ fiber optic probe. The measured measured temperatures temperatures were sustainable and reproducible probe. The within aa given givenset setofofexperiments experiments conditions (Figure These are consistent the much conditions (Figure 3). 3). These datadata are consistent with with the much more more detailed studies of Mingos [17], Berlan [19], and Chemat [18], which we did not endeavor to detailed studies of Mingos [17], Berlan [19], and Chemat [18], which we did not endeavor to duplicate duplicate in their entirety. in their entirety.

Figure 2.2.Temperature profiles of unstirred methanol, ethanol,ethanol, and isopropyl alcohol under microwave Figure Temperature profiles of unstirred methanol, and isopropyl alcohol under heating for 3 min. The accepted boiling point of each liquid is plotted for comparison. The measured microwave heating for 3 min. The accepted boiling point of each liquid is plotted for comparison. temperature oftemperature the liquid exceeded its normal boiling point inboiling each case. The measured of the liquid exceeded its normal point in each case.

The exact magnitude of the deviation (ΔT) between the measured temperature of the liquid and The exact magnitude of the deviation (∆T) between the measured temperature of the liquid and its normal boiling point depends on several factors including volume of the liquid, size and shape of its normal boiling point depends on several factors including volume of the liquid, size and shape the flask, applied microwave power, and additives such as boiling chips, stir bar, or even a fiber optic of the flask, applied microwave power, and additives such as boiling chips, stir bar, or even a fiber probe. Our specific experimental design produced measurable bulk liquid temperatures for methanol, optic probe. Our specific experimental design produced measurable bulk liquid temperatures for ethanol, and isopropyl alcohol that exceeded their normal boiling points by ΔT of 14 °C, 21 °C, and methanol, ethanol, and isopropyl alcohol that exceeded their normal boiling points by ∆T of 14 ˝ C, 28 °C above the expected values, respectively. Mingos previously measured analogous deviations of 21 ˝ C, and 28 ˝ C above the expected values, respectively. Mingos previously measured analogous ΔT = 19 °C, 24 °C, and 18 °C for the same three solvents, but using larger volumes and a multimode deviations of ∆T = 19 ˝ C, 24 ˝ C, and 18 ˝ C for the same three solvents, but using larger volumes and (domestic kitchen) microwave oven to heat the liquids [17]. Berlan observed a lesser degree of a multimode (domestic kitchen) microwave oven to heat the liquids [17]. Berlan observed a lesser superboiling, on the order of ΔT = 6 °C for methanol and 12 °C for ethanol, using smaller volumes degree of superboiling, on the order of ∆T = 6 ˝ C for methanol and 12 ˝ C for ethanol, using smaller (ca. 10 mL) of solvent and lower applied microwave power (ca. 40 W) [12]. Finally, Chemat reported superboiling of methanol and ethanol at ΔT = 14 °C and 11 °C [11], respectively, although these data 21675to an internal fiber optic probe. were measured by as external IR sensor as opposed 4

Molecules 2015, 20, 21672–21680

volumes (ca. 10 mL) of solvent and lower applied microwave power (ca. 40 W) [12]. Finally, Chemat reported superboiling of methanol and ethanol at ∆T = 14 ˝ C and 11 ˝ C [11], respectively, although Molecules 2015, 20, page–page these data were measured by as external IR sensor as opposed to an internal fiber optic probe.

Figure 3. 3. Temperature Figure Temperature profiles profiles of of unstirred unstirred isopropyl isopropyl alcohol alcohol(IPA) (IPA)under undermicrowave microwaveheating heatingfor for 10 min (3 trials). The accepted boiling point of IPA is plotted for comparison. 10 min (3 trials). The accepted boiling point of IPA is plotted for comparison.

Our interpretation of these collective data is that superboiling is a real but erratic phenomenon. Our interpretation of these collective data is that superboiling is a real but erratic phenomenon. It reliably occurs under microwave heating of unstirred liquids, but the exact magnitude is difficult It reliably occurs under microwave heating of unstirred liquids, but the exact magnitude is difficult to predict and/or reproduce, which makes it problematic. to predict and/or reproduce, which makes it problematic. Chemat described several methods for reducing the magnitude of “super-heated boiling” Chemat described several methods for reducing the magnitude of “super-heated boiling” (superboiling), including the addition of boiling stones, stirring, air bubble injection, sonication, and (superboiling), including the addition of boiling stones, stirring, air bubble injection, sonication, and even the use of an internal fiber optic probe. Each of these perturbations to the system reportedly even the use of an internal fiber optic probe. Each of these perturbations to the system reportedly produces additional nucleation sites at which the liquid→gas phase-change can occur, resulting in produces sites attemperature which the liquidÑgas phase-change occur,Chemat resulting in an an overalladditional regressionnucleation of the bulk liquid to the normal boiling point.can Of these, noted overall the bulk liquid temperature tosevere the normal boiling point. Of these, Chemat noted that theregression “action of of stirring or boiling stones is more and practically removes all super-heating.” that the “action of stirring or boiling stones is more severe and practically removes all super-heating.” In contrast, however, Kappe recently reported superboiling a toluene solution to ΔT of ca. 10 °C above In Kappe recently reported superboiling toluene to ∆Tare of ca. 10 ˝ Cinabove itscontrast, expectedhowever, boiling point “even applying vigorous stirring”a[14]. Oursolution observations largely line its expected boiling point “even applying vigorous stirring” [14]. Our observations are largely in line with those of Chemat, although the size of the stir bar plays a role, as follows. with those of Chemat, although the size of the stir bar plays a role, as follows. Our second series of experiments addresses the question of whether or not superboiling can Ourbe second series of experiments addresses thedetermined question ofthat whether or not was superboiling can reliably observed in stirred liquids. We previously superboiling not a factor reliably be observed in stirred liquids. We previously determined that superboiling was not a factor in our published experiments aimed at comparing reaction rates under conventional and microwave in our published experiments comparing rates under conventional and microwave heating (cf. Figure 1, above) aimed [9]. In at those studies,reaction our objective was to maintain consistent bulk heating (cf. Figure 1, conventionalabove) [9]. Inand those studies, our objective was to bulk temperature between microwave-heated experiments. Tomaintain that end, consistent we used a stir temperature between conventionaland microwave-heated experiments. To that end, we used a stir bar commensurate with the size of the reaction vessel, and we visually monitored the reflux behavior bar commensurate with the size of the reaction vessel, and we visually monitored the reflux behavior to ensure that boiling was qualitatively similar between the two heating methods. Here, our objective to that boiling was qualitatively between the twostirred heating Here,rate our allowed objective is ensure the opposite: to produce measurable similar superboiling in liquids atmethods. the maximum isbythe to produce measurable in liquids stirred at the rate× allowed theopposite: CEM microwave reactor. To thissuperboiling end, we chose a relatively small stirmaximum bar (10 mm 3 mm) by the CEM Tothe thisliquids end, we a relatively small stir bar (10 mm ˆ 3 mm) compared tomicrowave the volume reactor. (30 mL) of we chose were examining. compared to theinvolume mL) of the liquids were examining. As shown Figure (30 4, superboiling can be we observed in stirred methanol, ethanol, and isopropyl As shown in Figure 4, superboiling observedthe in magnitude stirred methanol, ethanol, andto isopropyl alcohol when using a relatively small stircan bar,bealthough of ΔT was reduced within ca. 5 °Cwhen of theusing normal boiling points. an extended look at isopropyl (Figure 5), alcohol a relatively smallWe stirthen bar, took although the magnitude of ∆T wasalcohol reduced to within which in which observed the greatest of ΔTatinisopropyl the absence of stirring (up5), ca. 5 ˝ Cisofthe theliquid normal boilingwe points. We then took anmagnitude extended look alcohol (Figure to 28 °C, Figures 2 and 3, above). saw occasional spikes in the measured temperature as stirring much which is the liquid in which we We observed the greatest magnitude of ∆T in the absencetoof ˝ as ΔT of ca. °C above the boiling point, butoccasional these maxima notmeasured sustained,temperature predictable,tooras (up to 28 C, 10 Figures 2 and 3 above). We saw spikeswere in the reproducible. On average, the measured ΔT for stirred isopropyl alcohol remained around 5 °C when using a relatively small stir bar in our experiments. 21676

Molecules 2015, 20, 21672–21680

much as ∆T of ca. 10 ˝ C above the boiling point, but these maxima were not sustained, predictable, or reproducible. On average, the measured ∆T for stirred isopropyl alcohol remained around 5 ˝ C when using relatively Molecules 2015,a 20, page–pagesmall stir bar in our experiments. Molecules 2015, 20, page–page

Figure 4. Temperature profiles of stirred methanol, ethanol, and isopropyl alcohol under microwave Figure Temperature profiles microwave Figure 4. 4. Temperature profiles of of stirred stirred methanol, methanol, ethanol, ethanol, and and isopropyl isopropyl alcohol alcohol under under microwave heating for 3 min. Liquids were stirred at the highest setting using a relatively small stir bar heating were stirred at the setting using using a relatively small stir bar (10 heating for for33min. min.Liquids Liquids were stirred at highest the highest setting a relatively small stir mm bar (10 mm × 3 mm). The accepted boiling point of each liquid is plotted as a straight horizon line for ˆ mm).× The accepted boiling point of each is plotted a straight for reference. (103 mm 3 mm). The accepted boiling pointliquid of each liquid isasplotted as ahorizon straightline horizon line for reference. The measured temperature (in situ fiber optic probe) of the liquid exceeded its normal The measured (in situ fiber (in optic probe) the liquid its normal boiling in reference. The temperature measured temperature situ fiber of optic probe)exceeded of the liquid exceeded its point normal boiling point in each case, although by less than what was observed in unstirred liquids. each case, although by less than what was observed in unstirred liquids. boiling point in each case, although by less than what was observed in unstirred liquids.

Figure 5. Temperature profiles for three experiments in which isopropyl alcohol (IPA) was subjected to microwave heating for 10 min, stirred the highest setting using a relatively small(IPA) stir bar (10subjected mm × 3 Figure 5. Temperature profiles for threeatexperiments experiments in which isopropyl alcohol was subjected Figure 5. Temperature profiles for three in which isopropyl alcohol (IPA) was mm). The accepted boiling point of isopropyl alcohol is plotted for comparison. The measured to microwave 10 min, stirred at theathighest setting setting using a relatively small stir bar (10 stir mmbar ×3 to microwaveheating heatingforfor 10 min, stirred the highest using a relatively small temperature exceeded its normal boiling point by about 5alcohol °C onfor with occasional and mm). The boiling pointboiling of isopropyl is plotted comparison. The measured (10 mm ˆ accepted 3 mm). The accepted point ofalcohol isopropyl isaverage, plotted for comparison. The unpredictable spikes inexceeded temperature to ca. 10 °C above theby boiling temperature exceeded its normal point bypoint about 5about °C point. on with with occasional and measured temperature itsboiling normal boiling 5 ˝average, C on average, occasional ˝ C above unpredictable spikes in temperature to ca.to10 the boiling point. and unpredictable spikes in temperature ca.°C 10above the boiling point.

When we switched to a larger stir bar (25 mm × 5 mm, Figure 6), the magnitude of ΔT dropped further to ca. °C of the normal boiling and the× occasional spikes longer When we2 switched to a larger stir point, bar (25 mm 5 mm, Figure 6), in thetemperature magnitudewere of ΔTnodropped 21677 observed, as2reported by Chemat. Finally, we note that boiling chips—in our case, made further to ca. °C of the previously normal boiling point, and the occasional spikes in temperature were no longer of groundas quartz glass—also reduced the magnitude ΔTnote to with °C ofchips—in the normalour boiling observed, reported previously by Chemat. Finally,ofwe that1–2 boiling case,point made (cf. Figure 1, above), whether used alone or in concert with stirring. of ground quartz glass—also reduced the magnitude of ΔT to with 1–2 °C of the normal boiling point

Molecules 2015, 20, 21672–21680

When we switched to a larger stir bar (25 mm ˆ 5 mm, Figure 6), the magnitude of ∆T dropped further to ca. 2 ˝ C of the normal boiling point, and the occasional spikes in temperature were no longer observed, as reported previously by Chemat. Finally, we note that boiling chips—in our case, made of ground quartz glass—also reduced the magnitude of ∆T to with 1–2 ˝ C of the normal boiling point (cf.2015, Figure 1, above), whether used alone or in concert with stirring. Molecules 20, page–page

Figure 6. 6. Temperature alcohol was was subjected subjected to to Figure Temperature profiles profiles for for three three experiments experiments in in which which isopropyl isopropyl alcohol microwave heating for 10 min, stirred at the highest setting using a relatively large stir bar (25 mm microwave heating for 10 min, stirred at the highest setting using a relatively large stir bar (25 mm× 5 mm). The accepted boiling measured ˆ 5 mm). The accepted boilingpoint pointofofisopropyl isopropylalcohol alcoholisis plotted plotted for for comparison. comparison. The The measured ˝ temperature exceeded its normal boiling point by about 2 °C on average. temperature exceeded its normal boiling point by about 2 C on average.

All of our temperature determinations for this superboiling study were made using a fiber optic All of our temperature determinations for this superboiling study were made using a fiber optic thermometer immersed in the liquid, so it was important to determine whether the presence of the fiber thermometer immersed in the liquid, so it was important to determine whether the presence of the optic was causing additional nucleation and yielding lower superboiling temperatures. To do this we fiber optic was causing additional nucleation and yielding lower superboiling temperatures. To do measured the internal temperature of the solution remotely using a thermal imaging camera. Thermal this we measured the internal temperature of the solution remotely using a thermal imaging camera. imaging cameras measure black-body radiation in the far infrared, typically between 8 and 12 μm. Thermal imaging cameras measure black-body radiation in the far infrared, typically between 8 and However, quartz or glass reaction vessels have optical cutoffs >4 μm. Therefore, we fabricated sample 12 µm. However, quartz or glass reaction vessels have optical cutoffs >4 µm. Therefore, we fabricated cells out of high-density polyethylene, which is transparent to infrared radiation in the range detected sample cells out of high-density polyethylene, which is transparent to infrared radiation in the range by the camera. The measured temperature of ≤85 °C from the thermal imaging camera (Figure 7) was detected by the camera. The measured temperature of ď85 ˝ C from the thermal imaging camera found to be in good agreement with the measured bulk temperature (≤85 °C) from the fiber optic probe, (Figure 7) was found to be in good agreement with the measured bulk temperature (ď85 ˝ C) from which is within about 2 °C of the expected boiling point for isopropyl alcohol (83 °C). In short, the the fiber optic probe, which is within about 2 ˝ C of the expected boiling point for isopropyl alcohol presence of the fiber optic probe did not appear to affect the boiling temperature. (83 ˝ C). In short, the presence of the fiber optic probe did not appear to affect the boiling temperature. Our interpretation is that one can achieve superboiling in stirred liquids, although not nearly to the same magnitude as in unstirred liquids, provided that a relatively small stir bar is used, and other potential nucleation sites are minimized. More importantly (e.g., for comparing rates of reactions in refluxing solvents heated conventionally vs. using microwave energy), one can mitigate superboiling by ensuring vigorous stirring with an appropriately sized stir bar. However, what constitutes “a relatively small stir bar” vs. “an appropriately sized stir bar” likely depends on the volume and identity of the liquid, size and shape of the flask, magnitude of applied microwave power, and other reaction variables that may not always be significant on a laboratory scale.

21678 Figure 7. Thermal image of the bulk liquid IPA while at reflux with stirring under microwave heating at 75 W of power.

However, quartz or glass reaction vessels have optical cutoffs >4 μm. Therefore, we fabricated sample cells out of high-density polyethylene, which is transparent to infrared radiation in the range detected by the camera. The measured temperature of ≤85 °C from the thermal imaging camera (Figure 7) was found to be in good agreement with the measured bulk temperature (≤85 °C) from the fiber optic probe, Molecules 20, 21672–21680 which is2015, within about 2 °C of the expected boiling point for isopropyl alcohol (83 °C). In short, the presence of the fiber optic probe did not appear to affect the boiling temperature.

Figure 7. Thermal image of the bulk liquid IPA while at reflux with stirring under microwave heating Figure 7. Thermal image of the bulk liquid IPA while at reflux with stirring under microwave heating at 75 W of power. at 75 W of power.

4. Conclusions Much of the early microwave chemistry literature has been called into question—sometimes 7 appropriately—due to concerns over irreproducibility and underestimation of bulk solution temperatures. Indeed, chemical reactions heated by microwave radiation are sensitive to variables that synthetic chemists are not necessarily accustomed to considering on a laboratory scale. In spite of this, dedicated microwave reactors have become standard equipment in many modern synthesis labs. Routine users benefit from careful and thoughtful assessments of thermal events that are or are not likely to be possible using microwave dielectric heating, but it may be difficult for the casual reader to separate “myth from reality” [20] when reading microwave chemistry literature. Based on our experiments, we conclude that the published literature accounts of microwave-assisted superheating and of the microwave-specific effect of nucleation-limited boiling (“superboiling”) are qualitatively correct: (1) temporary superheating of unstirred liquids is facilitated by microwave heating, although it can also be achieved with conventional heating; (2) temporary superheating of liquids persists under carefully controlled conditions only until nucleation is initiated, at which point boiling commences with rapid and sometimes violent release of excess thermal energy; (3) sustained superboiling of unstirred liquids can be observed thermometrically (and visually: superboiling more chaotic than conventional reflux) under the action of microwave heating, particularly in the absence of appropriate nucleation sites; (4) rapid stirring and/or boiling chips effectively mitigate microwave-specific superboiling; however; (5) one cannot presume to know the precise refluxing temperature of a bulk liquid without considering the dynamic interplay of factors including pressure, volume, solute identity and concentration, and the abundance and distribution of nucleation sites. Solvent reflux is commonly used in organic chemistry as a means of controlling bulk solution temperature under a given set of experiments conditions. It can also be used as a convenient means of identifying microwave-specific thermal effects, provided that one does not over-interpret the results in the absence of complementary thermometric data and control experiments. Acknowledgments: This work was supported under National Science Foundation CHE 1112046. Author Contributions: A.F. and J.H. performed the experiments and compiled the data. A.F. wrote a first draft of the manuscript. A.E.S. and G.B.D. guided the experimental design, supervised the experimental activities, and collaborated on the production of the final draft of the manuscript. Conflicts of Interest: The authors declare no conflict of interest.

21679

Molecules 2015, 20, 21672–21680

References and Notes 1. 2.

3. 4.

5. 6. 7. 8. 9. 10. 11.

12. 13. 14. 15. 16. 17. 18. 19. 20.

Rosana, M.R.; Tao, Y.; Stiegman, A.E.; Dudley, G.B. On the rational design of microwave-actuated organic reactions. Chem. Sci. 2012, 3, 1240–1244. [CrossRef] Rosana, M.R.; Hunt, J.; Ferrari, A.; Southworth, T.; Tao, Y.; Stiegman, A.E.; Dudley, G.B. Microwave-Specific Acceleration of a Friedel—Crafts Reaction: Evidence for Selective Heating in Homogeneous Solution. J. Org. Chem. 2014, 79, 7437–7450. [CrossRef] [PubMed] Chen, P.-K.; Rosana, M.R.; Dudley, G.B.; Stiegman, A.E. Parameters affecting the microwave-specific acceleration of a chemical reaction. J. Org. Chem. 2014, 79, 7425–7436. [CrossRef] [PubMed] Herrero, M.A.; Kremsner, J.M.; Kappe, C.O. Nonthermal Microwave Effects Revisited: On the Importance of Internal Temperature Monitoring and Agitation in Microwave Chemistry. J. Org. Chem. 2008, 73, 36–47. [CrossRef] [PubMed] Richards, E. Magical Microwaves. Available online: http://www.rsc.org/chemistryworld/News/2012/ February/microwave-effects-in-organic-reactions.asp (accessed on 2 February 2012). Ritter, S.K. Microwave Chemistry Remains Hot, Fast, and A Tad Mystical. Chem. Eng. News 2014, 92, 26–28. Ritter, S.K. More on Microwaves. Chem. Eng. News 2014, 92, 23. Kappe, C.O.; Pieber, B.; Dallinger, D. Microwave Effects in Organic Synthesis—Myth or Reality? Angew. Chem. Int. 2013, 52, 1088–1094. [CrossRef] [PubMed] Dudley, G.B.; Stiegman, A.E.; Rosana, M.R. Correspondence on microwave effects in organic synthesis. Angew. Chem. Int. 2013, 52, 7918–7923. [CrossRef] [PubMed] Kappe, C.O. Reply to the Correspondence on Microwave Effects in Organic Synthesis. Angew. Chem. Int. 2013, 52, 7924–7928. [CrossRef] [PubMed] King, A. Debate over Microwave Boost to Reactions Rumbles on. Available online: http://www.rsc. org/chemistryworld/2014/09/debate-over-microwave-specific-heating-rumbles (accessed on 10 September 2014). Dudley, G.B.; Richert, R.; Stiegman, A.E. On the Existence of and Mechanism for Microwave-Specific Reaction Rate Enhancement. Chem. Sci. 2015, 6, 2144–2152. [CrossRef] Kappe, C.O.; Dallinger, D.; Murphree, S.S. Practical Microwave Synthesis for Organic Chemists. Strategies, Instruments, and Protocols; Wiley–VCH: Weinheim, Germany, 2009; p. 32. Supporting information from [8], p 14. Baidakov, V.G. Explosive Boiling of Superheated Cryogenic Liquids; Wiley–VCH: Weinheim, Germany, 2009. Debenedetti, P.G. Metastable Liquids; Princeton University Press: Princeton, NJ, USA, 1996. Baghurst, D.R.; Mingos, D.M.P. Superheating Effects Associated with Microwave Dielectric Heating. J. Chem. Soc. Chem. Commun. 1992, 6, 674–677. [CrossRef] Chemat, F.; Esveld, E. Microwave Super-Heated Boiling of Organic Liquids: Origin, Effect, and Application. Chem. Eng. Technol. 2001, 24, 735–744. [CrossRef] Saillard, R.; Poux, M.; Berlan, J.; Audhuy-Peaudecerf, M. Microwave Heating of Organic Solvents: Thermal Effects and Field Modelling. Tetrahedron 1995, 51, 4033–4042. [CrossRef] Laurent, R.; Laporterie, A.; Dubac, J.; Berlan, J.; Lefeuvre, S.; Audhuy, M. Specific Activation by Microwaves: Myth or Reality? J. Org. Chem. 1992, 57, 7099–7102. [CrossRef] © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).

21680