Evaluation on the Photosensitivity of 2, 2-Azobis (2, 4-Dimethyl

molecules

Article Evaluation on the Photosensitivity of 2,20-Azobis(2,4-Dimethyl)Valeronitrile with UV

Yi Yang 1 and Yun-Ting Tsai 2,* ID

1 Shaanxi Key Laboratory of Prevention and Control of Coal Fire, School of Safety Science and Engineering, Xi’an University of Science and Technology, Xi’an 710054, China; [email protected] 2 School of Chemical Engineering and Technology, Xi’an Jiaotong University, Xi’an 710049, China * Correspondence: [email protected]; Tel.: +86-029-8266-5836

Received: 3 November 2017; Accepted: 12 December 2017; Published: 14 December 2017

Abstract: Azo compounds have high exothermic characteristics and low thermal stability, which have caused many serious thermal accidents around the world. In general, different locations (e.g., equatorial or polar regions) have different UV intensities. If the azo compound exists in an inappropriately stored or transported condition, the decrease in thermal stability may cause a thermal hazard or ageing. 2,20-Azobis(2,4-dimethyl)valeronitrile (ADVN) is investigated with respect to the thermal stability affected by UV exposure at 0, 6, 12, and 24 h. When ADVN is exposed to 24 h of 2 UV (100 mW/m and 254 nm), T0 is not only advanced, but the mass loss is also increased during the main decomposition stage. In addition, the apparent activation energy and integral procedural decomposition temperature (IPDT) of ADVN exposed to 24 h of UV is calculated by kinetic models. Therefore, the prevention mechanism, thermal characteristics, and kinetic parameters are established in our study. We should isolate UV contacting ADVN under any situations, avoiding ADVN being aged or leading to thermal runaway. This study provided signiﬁcant information for a safer process under changing UV exposure times for ADVN. Furthermore, the research method may serve as an important benchmark for handling potentially hazardous chemicals, such as azo compounds described herein.

Keywords: azo compound; free initiator; calorimetric and product analysis technology; kinetic parameters; environmental protection

1. Introduction In view of loss prevention, the appraisal of thermal characteristics for reactive chemicals has been a major concern of chemical industries. The main reason for many accidents is insufﬁcient knowledge of reactants, intermediates, products, or various catalysts. In general, a small change of temperature or pressure may sporadically cause a serious runaway reaction for highly-energetic chemicals. The most hazardous information can be obtained from the literature, but if this information is scanty, the related safety parameters should be assessed by proper thermal analysis technology [1–3]. 2,20-Azobis(2,4-dimethyl)valeronitrile (ADVN) (the chemical structure is shown in Figure1), a common azo compound, is an excellent free radical supplier for chemical processes. When azo is involved in an exothermic reaction, it can amply provide free radicals and energy for synthesis of organic compounds or polymers, such as styrene, methyl acrylate, epoxy resin, or propylene, as an initiator, cross-linking, or curing agent. However, ADVN has thermal instability and sensitivity because of its –N=N– structure, which may incur violent thermal decomposition by external ﬁre, other igniting sources, or irradiation [4–8]. In the past, many thermal accidents have occurred because the inherent safety of chemicals is not explicit and the accident process is unexpected, which may result in the stored temperature, cooling system, or pressure relief system to be wrongly designed

Molecules 2017, 22, 2219; doi:10.3390/molecules22122219 www.mdpi.com/journal/molecules Molecules 2017, 22, 2219 2 of 9 Molecules 2017,, 22,, 22192219 2 of 9 safetyfor safety considerations. considerations. Wanasekara Wanasekara et al. et in al. 2011 in 2011 indicated indicated that thatthe increase the increase in the in temperature the temperature and exposureand exposure time time may may lead lead to surface to surface degradation degradation and and embrittlement embrittlement of of fibers ﬁbers... WangWang Wang etet et al.al. al. inin in 2008 2008 mentioned that the UV light can cause the trans-cis isomerization mechanism of azo compounds (see Figure 22).). WangWang andand WangWang inin 2008 2008 alsoalso indicatedindicated thatthat thethe azoazo withwith aa transtrans isomerisomer hashas highhigh thermalthermal stability and low energy intensity [9–11]. [9–11]. Thus, Thus, it it is crucial that we understand the thermal hazard behaviors for azo which can be usedused efﬁcientlyefficiently andand safely duringduring operating,operating,operating, transportation,transportation, and and storage [12]. [12]. In In the the past, most of the research of ADVN ADVN for for runaway runaway excursion excursion has has been been the the analysis analysis of pure substances. However, the photo effect of ADVN has has rarely beenbeen explored.explored. Since Since ADVN ADVN has has extreme photosensitivity with UV, thethe decompositiondecompositionion mechanismmechanism maymay be changed changed with with the the exposureexposure timetime and and mass mass [13]. [[13].13]. Often,Often, UVUV UV irradiationirradiation irradiation existsexists exists ubiquitously,ubiquitously, ubiquitously, and and even even its its intensityintensity maymay bebe differentdifferent between locationlocation latitudes.latitudes. Simultaneously,Simultaneously, if if the the optical opticaltical isolation isolationisolation of ofof UV UVUV for forfor the thethe package packagepackage or oror container containercontainer is isnotis notnot complete, complete,complete, the the photothe photophoto effect effecteffect may decreasemaymay decreasedecrease the thermal thethe thermalthermal stability stabilitystability of ADVN, ofof resultingADVN,ADVN, resultingresulting in the increased inin thethe increasedriskincreased of a thermal riskrisk ofof a runawaya thermalthermal reaction runawayrunaway [14 reactionreaction]. [14].[14].

Figure 1. ChemicalChemical structure of ADVN.

Figure 2. LightLight effect effect for the trans-cis isomerization mechanism of ADVN.

The calorimetrycalorimetry andand kinetickinetic modelsmodels are are used used to to analyze analyze the the thermal thermal stability stability parameters, parameters, such such as asapparent apparent onset onset temperature temperature (T (0T),00),), peak peakpeak temperature temperaturetemperature (T ((Tp),pp),), ﬁnal finalfinal temperature temperaturetemperature ( ((TTfff),),), apparentapparent activation activation energy ( Eaa),),), frequencyfrequency frequency factorfactor (ln(ln k00),),), andand and reactionreaction reaction orderorder order (( (nn),),), toto establishestablish thethe hazardhazard characteristicscharacteristics and and toto explore explore thethe decompositiondecomposition products products by by employingemployemployinging differential differential scanningscanning calorimetrycalorimetry (DSC) (DSC) and and thermogravimetrythermogravimetry (TG) (TG) forfor determinationdeterminatiodetermination of UV effectseffects onon ADVNADVNADVN [[15,16].[15,16].15,16].

2. Results Results and and Discussion

2.1. TG Results Figure3 3 showsshows thethe massmass lossloss versusversus temperaturetemperature diagramdiagram fromfrom TGTG testingtesting forfor ADVNADVN andand ADVNADVN ◦ with different exposure durations of 6, 12 andand 2424 hh ofof UVUV irradiationirradiation atat aa heatingheating raterate 2020 °C/min.C/min. There areare two two decomposition decomposition stages stages for ADVN for ADVN and ADVN andand exposedADVNADVN toexposedexposed UV, and to theto UV,UV, main andand decomposition thethe mainmain decompositionstage is the second stage stage. is the We second found stage. that whenWe foun ADVNd that was when exposed ADVN to was 6 and exposed 12 h of to UV, 6 and the proﬁles12 h of of the TG curve are not obviously different between each other, and for both T is 78 ◦C. However, UV, the profiles of the TG curve are not obviously different between each other, and0 for both T00 isis 7878 °C.°C. although the T of the ﬁrst stage of ADVN is close to ADVN exposed to 24 h of UV, the T of the However, although0 the T00 ofof thethe firstfirst stagestage ofof ADVNADVN isis closeclose toto ADVNADVN exposedexposed toto 2424 hh ofof UV,UV,0 thethe T00 ◦ ◦ ofsecond the second decomposition decomposition stage is stage curtailed is curtailed from 132 fromC to 132 127 °CC to and 127 the °C massand the loss mass is increased loss is increased from 70% to 80%. According to the literature [17], T and mass loss can be used to determine the thermal stability fromfrom 70%70% toto 80%.80%. AccordingAccording toto thethe literatureliterature0 [17],[17], T00 andand massmass lossloss cancan bebe usedused toto determinedetermine thethe 2 thermalforthermal chemicals. stabilitystability Therefore, forfor chemicals.chemicals. if ADVN Therefore,Therefore, is exposed ifif ADVN toADVN a UV isis intensity exposedexposed of toto 100 aa UVUV mW/m intensityintensityfor over ofof 100100 24 mW/m h,mW/m the onset22 forfor over 24 h, the onset temperature and temperature of the maximum mass loss can decrease, causing abnormal aging or lessening the thermal stability during preparation, storage, transportation, and operating conditions.

Molecules 2017, 22, 2219 3 of 9 temperature and temperature of the maximum mass loss can decrease, causing abnormal aging or Molecules 2017, 22, 2219 3 of 9 lesseningMolecules 2017 the, 22 thermal, 2219 stability during preparation, storage, transportation, and operating conditions.3 of 9 ADVN (Without UV) ADVNADVN with(Without UV-6.0 UV) hr ADVN with UV-6.0 hr 100 ADVN withUV-12.0 hr ADVN with UV-12.0 hr o ADVNADVN withwith UV-24.0UV-12.0 hrhr 100 TADVN: 78.9 withUV-12.0 C hr 01 o ADVN with UV-24.0 hr T : 78.9 oC T 01: 133.5 C 80 02 T : 133.5 oC 80 02

60 ADVN with UV-24.0 hr 60 TADVN: 77.9 with oC UV-24.0 hr ADVN with UV-6.0 hr 01 o T : 77.9 oC ADVN witho UV-6.0 hr T 01: 127.3 C 40 T : 78.7 C 02 o 01 o T : 127.3 C 40 o 02 Mass loss (%) T : 78.7 C T 01: 132.2 C 02 o Mass loss (%) T : 132.2 C 20 02 ADVN (No UV) 20 TADVN: 78.6 (No oC UV) 01 o T : 78.6 oC T 01: 133.1 C 0 02 T : 133.1 oC 0 02 60 80 100 120 140 160 180 200 60 80 100 120 140 160 180 200 Temperature (oC) o Temperature ( C) Figure 3. TG thermal curve of ADVN and ADVN with different exposure durations of UV/6, 12 and 24 h. Figure 3. TGTG thermal thermal curve curve of of ADVN ADVN and ADVN with differentdifferent exposure durations ofof UV/6,UV/6, 12 12 and and 24 24 h. In Figure 4, we compared the IPDT for ADVN and ADVN with different exposure durations of In Figure 44,, wewe comparedcompared the IPDTIPDT forfor ADVN ADVN and and ADVN ADVN with with different different exposure exposure durations durations of 6, 12 and 24 h of UV irradiation, corresponding to values of 119, 125, 124 and 112, respectively. The of6, 12 6, 12and and 24 h 24 of h UV of UVirradiation, irradiation, corresponding corresponding to values to values of 119, of 119,125, 125,124 and 124 112, and respectively. 112, respectively. The results showed that when the exposure time of UV is 6 and 12 h, the IPDT could be increased, but the Theresults results showed showed that thatwhen when the exposure the exposure time time of UV of UVis 6 isand 6 and12 h, 12 the h, IPDT the IPDT couldcould be increased, be increased, but butthe IPDT of ADVN is decreased after exposure to UV for 24 h. According to the literature, the thermal theIPDTIPDT of ADVNof ADVN is decreased is decreased after after exposure exposure to toUV UV fo forr 24 24 h. h. According According to to the the literature, literature, the thermal characteristics of chemicals are determined by energy and states of chemical bonds [18]. UV light can characteristics of chemicals are determined byby energyenergy andand statesstates ofof chemical bondsbonds [[18].18]. UV light can lead to the formation of the trans-cis isomerization of ADVN, causing a decrease in the thermal lead toto the the formation formation of of the the trans-cis trans-cis isomerization isomerization of ADVN, of ADVN, causing causing a decrease a decrease in the thermal in the stability.thermal stability. In general, 24 h is usually considered as a standard emergent time or transportation for Instability. general, In 24general, h is usually 24 h is considered usually considered as a standard as a emergentstandard emergent time or transportation time or transportation for chemical for chemical industries, so that ADVN should avoid being in contact with UV irradiation under any industries,chemical industries, so that ADVN so that should ADVN avoid should being avoid in contact being within contact UV irradiation with UV irradiation under any conditions.under any conditions. The TG and IPDT results for ADVN exposed to 6, 12, and 24 h of UV at a heating rate Theconditions. TG and TheIPDT TGresults and IPDT for ADVN results exposed for ADVN to 6, exposed 12, and 24 to h 6, of 12, UV and at a24 heating h of UV rate at 20 a ◦heatingC are listed rate 20 °C are listed in Table 1. in20 Table°C are1. listed in Table 1.

Table 1. Thermal stability parametersparameters bybyTG TG testing testing for for ADVN ADVN and and ADVN ADVN exposed exposed to to 6, 6, 12 12 and and 24 24 h ofh Table 1. Thermal stability parameters by TG testing for ADVN and ADVN exposed to 6, 12 and 24 h ofUV UV at at a heatinga heating rate rate 20 20◦C. °C. of UV at a heating rate 20 °C. Mass Loss Mass Loss Sample T01 (°C) T02 (°C) Mass Loss Mass Loss IPDT/°C Sample T (◦C) T (◦C) (FirstMass Stage) Loss (%) (SecondMass Stage) Loss (%) IPDT/◦C Sample T01 (°C)01 T02 (°C) 02 (First Stage) (%) (Second Stage) (%) IPDT/°C ADVN/without UV 78.6 132.1 (First 30.0Stage) (%) (Second 70.0 Stage) (%) 119.0 ADVN/without UV 78.6 132.1 30.0 70.0 119.0 ADVN/withoutADVN/6 h of UV UV 78.7 78.6 133.2 132.1 30.0 30.0 70.0 70.0 125.0 119.0 ADVN/6ADVN/6 h of UV h of UV 78.7 78.7 133.2 133.2 30.0 30.0 70.0 70.0 125.0 125.0 ADVN/12ADVN/12 h of UV h of UV 78.9 78.9 133.5 133.5 25.0 25.0 75.0 75.0 124 124 ADVN/24ADVN/12ADVN/24 hh ofof UVUV h of UV 77.9 78.9 77.9 127.3 133.5 127.3 20.0 25.0 20.0 80.0 80.0 75.0 112.0 112.0 124 ADVN/24 h of UV 77.9 127.3 20.0 80.0 112.0

130 ADVN/24.0 hr UV 130 ADVN/12ADVN/24..00 hr hr UVUV 128 ADVN/6.0 hr UV ADVN/6ADVN/12.0. hr0 hr UV UV 128 ADVN/6 0 hr oUV 126 IPDT = 125. .0 C ADVN/6 0 hr UV o Pure ADVN. 126 IPDT = 125.0 C 124 Pure ADVN ADVN/12 hr UV 124 122 IPDTADVN/12 = 124 hr.0 oUVC 122 o 120 IPDT = 124.0 C High C)

o 120 High

( 118 Low C)

o Pure ADVN

( 118 o Low 116 IPDTPure ADVN= 119.0 C IPDT 116 IPDT = 119.0 oC

IPDT 114 114 112 ADVN/24 hr UV 112 110 IPDTADVN/24 = 112 hr.0 oUVC 110 o 108 IPDT = 112.0 C 108 106 106 0 6 12 18 24 0 6Exposure time 12 of UV/hr 18 24 Exposure time of UV/hr FigureFigure 4. Calculation of IPDTIPDT forfor ADVN with different exposure durations of UV/6, 12, 12, and and 24 24 h h by by Figure 4. Calculation of IPDT for ADVN with different exposure durations of UV/6, 12, and 24 h by TG testing. TG testing. Based on the above-mentioned results, the following experiments focused on the UV effect for Based on the above-mentioned results, the following experiments focused on the UV effect for ADVN with an exposure time of 24 h. ADVN exposed to UV for 24 h was tested by TG at heating ADVN with an exposure time of 24 h. ADVN exposed to UV for 24 h was tested by TG at heating rates of 1, 5, 15 and 20 °C, as delineated in Figure 5. We also compared the IPDT for the results of four rates of 1, 5, 15 and 20 °C, as delineated in Figure 5. We also compared the IPDT for the results of four

Molecules 2017, 22, 2219 4 of 9

Based on the above-mentioned results, the following experiments focused on the UV effect for ADVN with an exposure time of 24 h. ADVN exposed to UV for 24 h was tested by TG at heating Molecules 2017, 22, 2219 4 of 9 Moleculesrates of 2017 1, 5,, 22 15, 2219 and 20 ◦C, as delineated in Figure5. We also compared the IPDT for the results of4 four of 9 ◦ ◦ heatingheating ratesrates andand thethe valuevalue ofof IPDTIPDT waswas increasedincreased fromfrom 101101 °CC toto 119119 °CC withwith anan increasingincreasing heatingheating heating rates and the value of IPDT was increased from 101 °C to 119 °C with an increasing heating2 rate.rate. Moreover,Moreover, we we plotted plotted the theIPDT IPDTversus versus the the heating heating rate rate curve, curve, and and there there was was an an extremely extremely high highR rate. Moreover, we plotted the IPDT versus the heating rate curve, and there was an extremely high valueR2 value from from the the linear linear regression. regression. According According to the to the results, results, the the heating heating rate rate can can determine determine the IPDTthe IPDTfor R2 value from the linear regression. According to the results, the heating rate can determine the IPDT thefor the decomposition decomposition reaction reaction of ADVNof ADVN exposed exposed to to UV UV for for 24 24 h. h. From From the the viewpoint viewpoint of of safersafer processprocess for the decomposition reaction of ADVN exposed to UV for 24 h. From the viewpoint of safer process design,design, whenwhen the the operating operating conditions conditions of of a chemicala chemic processal process involving involving ADVN ADVN exposed exposed to UV to forUV 24 for h design, when the operating conditions of a chemical process involving ADVN exposed to UV for need24 h need a low a reaction low reaction temperature, temperature, a low a heating low heating rate should rate should be proper be proper and reliable. and reliable. However, However, a high a 24 h need a low reaction temperature, a low heating rate should be proper and reliable. However, a heatinghigh heating rate is rate better is better for a highfor a reactionhigh reaction temperature. temperature.IPDT IPDTwill bewill used be used as a basic,as a basic, rudimentary, rudimentary, and high heating rate is better for a high reaction temperature. IPDT will be used as a basic, rudimentary, dependableand dependable thermal thermal stability stability parameter parameter for future for study.future Thestudy. linear The dependence linear dependence of IPDT withof IPDT different with and dependable thermal stability parameter for future study. The linear dependence of IPDT with heatingdifferent rates heating is depicted rates is indepicted Figure6 in. Figure 6. different heating rates is depicted in Figure 6.

24 hr UV/1.0 oC/min o o o 24 24 hr hr UV/1.0 UV/5.0 C/minC/min 24.0 hr UV/5.0 C/min o o o 100 24.0 hr UV/5.0o C/min 24 24 hr hr UV/5.0 UV/10.0 C/min C/min T : 72.0 C o 01 o 24 hr UV/10.0 C/mino 100 T : 72.0 Co 24 hr UV/20.0 C/min T01 : 158.0 C o 02 24 hr UV/20.0 C/min T : 158.0 oC 80 02 80 24.0 hr UV/10 oC/min o 24.0T : hr77.0 UV/10 oC C/min 60 01 o T : 77.0 Co T01 : 162.0 C 60 02 T : 162.0 oC 02 40 o Mass loss/% Mass 40 24.0 hr UV/20.0 C/min o o Mass loss/% Mass 24.0T : hr77.0 UV/20.0 C C/min 20 01 o T : 77.0 Co 20 o T01 : 175.0 C 24.0 hr UV/1.0 C/min 02 o o o T : 175.0 C 24.0 hr UV/1.0 C/min 02 T01: 55.0 C 0 o T : 55.0 Co T01 : 148.0 C 0 02 T : 148.0 oC 002 50 100 150 200 250 300 0 50 100 150 200 250 300 Temperature (oC) o Temperature ( C) Figure 5. TG thermal curve of ADVN with UV/24 h at heating rates of 1, 5, 10, and 20◦ °C/min. FigureFigure 5.5. TGTG thermalthermal curvecurve ofof ADVNADVN withwith UV/24UV/24 h at heating rates of 1, 5, 10, and 20 °C/min.C/min.

24.0 hr/20.0 oC/min 24.0 hr/20.0 oC/min T0 = 77.0 120 T = 77.0 120 T0 = 175.0 f T = 175.0 118 IPDTf : 109.0 oC 118 o o IPDT: 109.0 C 24.0 hr/1.0 C/min 116 o 24.0 hr/1.0 C/min o 116 T0 = 55.0 24 hr/10 C/min T = 55.0 o 114 0 24T hr/10= 77 C/min Tf = 150.0 0 114 T = 150.0 T = 77 f o T0 = 162 C) f o IPDT: 148.0 C 112 o

( T = 162 C) f o o 112 IPDT: 148.0 C IPDT: 119 C ( o 110 IPDT: 119 C IPDT 110 o IPDT 24.0 hr/5.0 C/min o 108 24.0T = hr/5.072.0 C/min 0 108 T = 72.0 T0 = 158.0 106 f T = 158.0 106 IPDTf : 113.0 oC o 104 IPDT: 113.0 C 104 0 5 10 15 20 0 5 10 15 20 Heating rate (oC/min) o Heating rate ( C/min) Figure 6. Linear dependence of IPDT for ADVN at 24 h UV at heating rates of 1, 5, 10, and 20 °C/min. Figure 6. LinearLinear dependence of IPDT for ADVN atat 2424 hh UVUV atat heatingheating ratesrates ofof 1,1, 5,5, 10,10, andand 2020 ◦°C/min.C/min. Table 2. Reaction kinetic simulation for ADVN with UV/24 h using non-linear regression methods at Table 2. Reaction kinetic simulation for ADVN with UV/24 h using non-linear regression methods at 2.2. Determinationheating rates of of 1, Kinetic 2, 4, and Models 8 °C/min. for ADVN With UV Irradiation heating rates of 1, 2, 4, and 8 °C/min. To investigateHeating Rate the reaction mechanism and predict the kinetic behaviors for ADVN exposed to Heating Rate Ea (kJ/mol) ln (A) (1/s) Reaction Order (n) ΔHd (kJ/kg) 24 h of UV,(°C/min) non-isothermal E kinetica (kJ/mol) analysis was ln (A conducted) (1/s) based Reaction on Order DSCdata, (n) whichΔHd is(kJ/kg) illustrated in (°C/min) Figure7. Followed1.0 ICTAC recommendations,137.0 the39.0 kinetic calculation 1.4 requests three up 632.0 temperature 1.0 137.0 39.0 1.4 632.0 programs. The2.0 complex reaction144.0 model can use the42.0 non-linear model-ﬁtting 1.5 method to 632.0 test the kinetic 2.0 144.0 42.0 1.5 632.0 4.0 140.0 40.0 1.4 603.0 4.0 140.0 40.0 1.4 603.0 8.0 138.0 39.0 1.5 633.0 8.0 138.0 39.0 1.5 633.0

Molecules 2017, 22, 2219 5 of 9 Molecules 2017, 22, 2219 5 of 9 2.2. Determination of Kinetic Models for ADVN With UV Irradiation 2.2. Determination of Kinetic Models for ADVN With UV Irradiation To investigate the reaction mechanism and predict the kinetic behaviors for ADVN exposed to To investigate the reaction mechanism and predict the kinetic behaviors for ADVN exposed to 24 h of UV, non-isothermal kinetic analysis was conducted based on DSC data, which is illustrated Molecules24 h of UV,2017, 22non-isothermal, 2219 kinetic analysis was conducted based on DSC data, which is illustrated5 of 9 in Figure 7. Followed ICTAC recommendations, the kinetic calculation requests three up temperature in Figure 7. Followed ICTAC recommendations, the kinetic calculation requests three up temperature programs. The complex reaction model can use the non-linear model-fitting method to test the kinetic programs. The complex reaction model can use the non-linear model-fitting method to test the kinetic parameters [19]. [19]. Figure Figure 8 8displays displays (a) (a) heat heat production production versus versus time time and (b) and heat (b) production heat production rate versus rate parameters [19]. Figure 8 displays (a) heat production versus time and (b) heat production rate versus timeversus by time experiments by experiments and simulations. and simulations. From From the thesimulation simulation results, results, we we observed observed that that there there was time by experiments and simulations. From the simulation results, we observed that there was significantsigniﬁcant curve ﬁttingfitting and nearly the same kinetickinetic parameters,parameters, includingincluding nn,, EEaa,, A,, andand ∆ΔHd,, for for both both significant curve fitting and nearly the same kinetic parameters, including n, Ea, A, and ΔHd, for both plots of heat production and heat production rate versus time of ADVN exposed to to 24 24 h of UV at plots of heat production and heat production rate versus time of ADVN exposed to 24 h of UV at heating rates ofof 1,1, 2,2, 44 andand 8 8◦ °C/minC/min byby simplesimplen nthth order order kinetic kinetic models. models. Therefore, Therefore, ADVN ADVN exposed exposed to heating rates of 1, 2, 4 and 8 °C/min by simple nth order kinetic models. Therefore, ADVN exposed to24 24 h ofh of UV UV is is determined determined as as an ann thnth order order reaction, reaction, instead instead of of autocatalysis, autocatalysis, andand thethe obtainedobtained kinetic to 24 h of UV is determined as an nth order reaction, instead of autocatalysis, and the obtained kinetic parameters can be used to predict other heating rate ratess or reaction temperatures for reliable information parameters can be used to predict other heating rates or reaction temperatures for reliable information inin the the future studies. The The related related kineti kineticc parameters are givengiven in Table2 2[20 [20].]. in the future studies. The related kinetic parameters are given in Table 2 [20].

o 24.0 hr UV/4.0 C/min 4.0 o o 4.0 o 24.0 hr UV/4.0o C/min 24.0 hr UV/1.0 C/min 24 hr UV/2.0 C/min T : 100.0 C o o 01 o 24.0 hr UV/1.0 o C/min 24 hr UV/2.0o C/min T : 100.0 C 24.0 hr UV/2.0 C/min 3.5 T : 86.0 C ΔH01: 602.0 J/g o 01 o d 24.0 hr UV/2.0 o C/min 3.5 T : 86.0 C ΔH : 602.0 J/g 24.0 hr UV/4.0 C/min ΔH01: 632.0 J/g d o d 24.0 hr UV/4.0 o C/min 3.0 ΔH : 632.0 J/g 24.0 hr UV/8.0 C/min 3.0 d 24.0 hr UV/8.0 oC/min 2.5

2.5 o o 24.0 hr UV/8.0 C/min 24.0 hr UV/1.0 C/min o 2.0 o 24.0 hr UV/8.0o C/min 24.0 hr UV/1.0o C/min T : 108.0 C 2.0 T : 67.0 C 01 o 01 o T : 108.0 C T : 67.0 C ΔH01: 632.0 J/g 1.5 d ΔH01 : 628.0 J/g d ΔH : 632.0 J/g 1.5 d ΔH : 628.0 J/g

Heat flow (W/g) Heat flow d

Heat flow (W/g) Heat flow 1.0 1.0 0.5 0.5 0.0 0.0 60 70 80 90 100 110 120 130 140 150 60 70 80 90 100 110 120 130 140 150 Temperature (oC) o Temperature ( C) Figure 7. DSC thermal curves of ADVN with 24 h UV at heating rates of 1, 2, 4 and 8 °C/min.◦ FigureFigure 7. DSC 7. DSC thermal thermal curves curves of ADVN of ADVN with with 24 h 24 UV h UVat heating at heating rates rates of 1, of 2, 1, 4 2,and 4 and 8 °C/min. 8 C/min.

(a) (b) (a) 700 (b) 700 3.0 1 oC/min/Exp. 600 3.0 o 1o C/min/Exp. 600 1 C/min/Sim. 2.5 o 2 1 o C/min/Exp.C/min/Sim. 500 2.5 o 500 2o C/min/Exp. ) 2 C/min/Sim. s) o o ) 2.0 2 C/min/Sim. 400 s) 4 C/min/Exp. o 2.0 o 400 1 C/min/Exp. 4o C/min/Exp. o 4 C/min/Sim. 1o C/min/Exp. o 1 C/min/Sim. 1.5 4o C/min/Sim. 300 o 8 C/min/Exp. 1o C/min/Sim. o 300 2 C/min/Exp. 1.5 8o C/min/Exp. o 8 C/min/Sim. o 200 2 2 C/min/Sim.C/min/Exp. 8 oC/min/Sim. o 1.0 o 200 4 2 C/min/Exp.C/min/Sim. 1.0 o o Heat Production (kJ/kg 100 4 4 C/min/Sim.C/min/Exp.

o Heat Production (kJ/kg 0.5 Heat Production (kJ/kg 100 4o C/min/Sim.

8 C/min/Exp. Heat Production (kJ/kg 0.5 o 0 8 8 o C/min/Sim.C/min/Exp. 0 8 oC/min/Sim. 0.0 -100 0.0 -100-500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 -500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 -500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 -500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 Time (sec) Time (sec) Time (sec) Time (sec) Figure 8. (a) Heat production versus time and (b) heat production rate versus time of model fitting Figure 8. (a) Heat production versus time and (b) heat production rate versus time of model fitting withFigure non-isothermal 8. (a) Heat production DSC at versusheatin timeg rates and of (b 1,) heat 2, 4, production and 8 °C/min rate versus for ADVN time of with model 24 ﬁttingh UV with via with non-isothermal DSC at heating rates of 1, 2,◦ 4, and 8 °C/min for ADVN with 24 h UV via experimentsnon-isothermal and DSC simulations. at heating rates of 1, 2, 4, and 8 C/min for ADVN with 24 h UV via experiments andexperiments simulations. and simulations. 3. Materials and Methods 3. MaterialsTable 2. Reactionand Methods kinetic simulation for ADVN with UV/24 h using non-linear regression methods at ◦ 3.1. Sampleheating rates of 1, 2, 4, and 8 C/min. 3.1. Sample The sample chosen◦ was ADVN 98 wt %, which was purchased from ACE Chemical Corp., HeatingThe sample Rate (chosenC/min) was EADVNa (kJ/mol) 98 wt ln%, (Awhich) (1/s) was Reaction purchased Order from (n) ACE∆ HChemicald (kJ/kg) Corp., Taoyuan, Taiwan. ADVN is light sensitive and thermally unstable, so it should be maintained in a Taoyuan, Taiwan.1.0 ADVN is light 137.0 sensitive and 39.0thermally unstable, 1.4 so it should be maintained 632.0 in a dark space and at a low temperature of 4 °C. The UV instrument was purchased from Uvitron dark space and2.0 at a low temperature 144.0 of 4 °C. 42.0The UV instrument 1.5 was purchased 632.0 from Uvitron International, 4.0Inc. (West Springfield, 140.0 MA, USA). 40.0 The exposure time 1.4 and intensity of UV 603.0 irradiation International, Inc. (West Springfield, MA, USA). The exposure time and intensity of UV irradiation is 6, 12 and 24 8.0h, respectively, to determine 138.0 the UV 39.0 effect on ADVN.1.5 The intensities/irradiance 633.0 of their is 6, 12 and 24 h, respectively, to determine the UV effect on ADVN. The intensities/irradiance of their

Molecules 2017, 22, 2219 6 of 9

3. Materials and Methods

3.1. Sample The sample chosen was ADVN 98 wt %, which was purchased from ACE Chemical Corp., Taoyuan, Taiwan. ADVN is light sensitive and thermally unstable, so it should be maintained in a dark space and at a low temperature of 4 ◦C. The UV instrument was purchased from Uvitron International, Inc. (West Springﬁeld, MA, USA). The exposure time and intensity of UV irradiation is 6, 12 and 24 h, respectively, to determine the UV effect on ADVN. The intensities/irradiance of their UV light exposures is 100 W/m2 and 254 nm. This value is the average intensity of UV at noon during the summer in China.

3.2. Differential Scanning Calorimetry (DSC) Temperature-controlled thermal curves of DSC experiments facilitate understanding the exothermic or endothermic reaction of a chemical, such as crystallization, curing reaction, phase change, or thermal decomposition reaction. Thus, it can be used as a safety assessment methodology to provide the thermal hazard information for ADVN. The type of DSC is selected in the Mettler DSC 821e. To investigate thermal characteristics of ADVN with UV irradiation, we heated it from 30 ◦C to 300 ◦C at different heating rates, here, 1, 2, 4 and 8 ◦C/min, using the DSC test. The sample amount for each experiment was approximately 1.5–5 mg and the sample was sealed in a gold-plated crucible [21].

3.3. Thermogravimetry (TG) Thermogravimetry, using a PerkinElmer Clarus 680 unit, was used to analyze the thermal decomposition products for ADVN and ADVN exposed to UV with different times of 6, 12 and 24 h. The TG experiments were performed from ambient temperature to 300 ◦C with heating rates of 1, 5, 10 and 20 ◦C/min in nitrogen gas purged at a ﬂow rate of 100 mL/min [22].

3.4. Evaluation of Integral Procedural Decomposition Temperature (IPDT) An evaluation parameter of thermal stability, IPDT, was used to consider and explore the overall thermal stability during the decomposition process. Since thermal stability involves three crucial factors of chemicals, including initial reaction, end reaction, and ratio for mass loss, IPDT was created based on different thermogravimetric regions of the TG curves. The equations of IPDT are as follows [23,24]:

◦ ◦ IPDT = A × K × Tf − Ti + Ti (1)

A◦ = (S1 + S2)/(S1 + S2 + S3) (2)

K◦ = (S1 + S2)/(S1) (3) ◦ ◦ where A and K is the proportion of mass loss. Ti and Tf are the mass loss at the initial temperature and ﬁnal temperature, respectively. S1, S2, and S3 are different partitioned regions. S1 is the thermogravimetric area under the TG curve; S2 is the thermogravimetric area of the non-mass loss; and S3 is the thermogravimetric area above the TG curve. The schematic diagram of IPDT is shown in Figure9. Molecules 2017, 22, 2219 6 of 9

UV light exposures is 100 W/m2 and 254 nm. This value is the average intensity of UV at noon during the summer in China.

3.2. Differential Scanning Calorimetry (DSC) Temperature-controlled thermal curves of DSC experiments facilitate understanding the exothermic or endothermic reaction of a chemical, such as crystallization, curing reaction, phase change, or thermal decomposition reaction. Thus, it can be used as a safety assessment methodology to provide the thermal hazard information for ADVN. The type of DSC is selected in the Mettler DSC 821e. To investigate thermal characteristics of ADVN with UV irradiation, we heated it from 30 °C to 300 °C at different heating rates, here, 1, 2, 4 and 8 °C/min, using the DSC test. The sample amount for each experiment was approximately 1.5–5 mg and the sample was sealed in a gold-plated crucible [21].

3.3. Thermogravimetry (TG) Thermogravimetry, using a PerkinElmer Clarus 680 unit (company, city, state abbrev. if USA, country), was used to analyze the thermal decomposition products for ADVN and ADVN exposed to UV with different times of 6, 12 and 24 h. The TG experiments were performed from ambient temperature to 300 °C with heating rates of 1, 5, 10 and 20 °C/min in nitrogen gas purged at a flow rate of 100 mL/min [22].

= °× °×( − )+ IPDT A K Tf Ti Ti (1)

A° = ()()S1+ S2 / S1+ S2+ S3 (2)

K ° = ()()S1 + S 2 / S1 (3)

where A° and K° is the proportion of mass loss. Ti and Tf are the mass loss at the initial temperature and final temperature, respectively. S1, S2, and S3 are different partitioned regions. S1 is the thermogravimetric area under the TG curve; S2 is the thermogravimetric area of the non-mass loss; Moleculesand S3 is2017 the, 22 thermogravimetric, 2219 area above the TG curve. The schematic diagram of IPDT is shown7 of 9 in Figure 9.

100

S1 S3 60

T T i f 40 Mass loss (%) Mass

S2 0

0 50 100 150 200 250 300 350 o Temperature ( C) Figure 9. Schematic diagram for the conception of IPDT [23,24]. Figure 9. Schematic diagram for the conception of IPDT [23,24].

3.5. Non-Linear Regression Method to Decide the Reaction Kinetics of ADVN/UV Irradiation In general, typical fundamental reactions of azo are nth order and autocatalytic; they may be composed of either a one-stage or a multi-stage reaction. However, linear regression methods, such as Kissinger, Ozawa, or Friedman models, are unable to predict complex chemical reactions of azo. Therefore, the non-linear regression method is usually used to directly ﬁt the thermal curve of chemicals through kinetic models to appraise the chemical reaction stages and various kinetic parameters, such as n, Ea, A, and the autocatalytic constant (z) for use in chemical engineering processes. The nth order and autocatalytic reaction models are presented in Equations (4) and (5), correspondingly [25–29]:

dα −E = A exp a (1 − α)n (4) dt RT

dα −E = A exp a (1 − α)n1 (αn2 + z) (5) dt RT where n1 and n2 are reaction orders for the two different reaction stages, α is the degree of conversion, and z is the autocatalytic constant.

4. Conclusions We tested the thermal stability of ADVN exposed to UV at 100 W/m2 and 254 nm, which is the average intensity of UV at noon during the summer in China. Based on the experimental results, when ADVN is exposed for 6 h and 12 h, the IPDT is decreased, demonstrating that the short exposure time cannot affect the thermal stability for ADVN. However, when ADVN is exposed to 24 h of UV, T0 is not only advanced, but the mass loss can also increase during the main decomposition stage. UV light can cause ADVN to the trans-cis isomerization, causing the decrease in the thermal stability. We should isolate UV contacting ADVN under any situation, avoiding ADVN being aged or leading to thermal runaway. In addition, IPDT and non-isothermal kinetic models are carried out to evaluate the thermal stability, reaction mechanism, and related kinetic parameters of ADVN exposed to UV irradiation. From the simulation results, we observed that there was significant curve fitting and nearly the same kinetic parameters. ADVN exposed to 24 h of UV is determined as an nth order reaction, instead of autocatalysis, and the obtained kinetic parameters can be used to predict other heating rates or reaction temperatures for reliable information in future studies. This study provided significant information on a safer process under changing exposure time of UV for ADVN. Furthermore, the research method may serve as an important benchmark for handling potentially hazardous chemicals, such as azo compounds described herein. Molecules 2017, 22, 2219 8 of 9

Acknowledgments: Financial support for this work was provided by projects of the China Postdoctoral Science Foundation (No. 2017M623181). The technical supports for this work from Chi-Min Shu worked at National Yunlin University of Science and Technology in Taiwan are also gratefully acknowledged. Author Contributions: Yun-Ting Tsai convinced and designed the project; Yi Yang wrote this manuscript. All authors read and approved the final manuscript. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Wang, Y.W.; Duh, Y.S.; Shu, C.M. Evaluation of adiabatic runaway reaction and vent sizing for emergency relief from DSC. J. Therm. Anal. Calorim. 2006, 85, 225–234. [CrossRef] 2. You, M.L. Thermal hazard evaluation of cumene hydroperoxide-metal ion mixture using DSC, TAM III, and GC/MS. Molecules 2016, 21, 562. [CrossRef] 3. Werber, J.; Wang, Y.J.; Milligan, M.; Li, X.; Ji, J.A. Analysis of 2,2’-azobis (2-amidinopropane) dihydrochloride degradation and hydrolysis in aqueous solutions. J. Pharm. Sci. 2013, 100, 3007–3015. [CrossRef][PubMed] 4. Chen, W.T.; Chen, W.C.; You, M.L.; Tsai, Y.T.; Shu, C.M. Evaluation of thermal decomposition phenomenon for 1,1-bis(tert-butylperoxy)-3,3,5-trimethylcyclohexane by DSC and VSP2. J. Therm. Anal. Calorim. 2015, 122, 1125–1133. [CrossRef] 5. Li, K.Y.; Tsai, S.Y.; Lin, C.P.; Tsai, Y.Y.; Shu, C.M. Smart technology for evaluating ﬁre extinguishing effect of tert-butyl hydroperoxide. Ind. Eng. Chem. Res. 2013, 52, 10969–10976. [CrossRef] 6. Tong, J.W.; Chen, W.C.; Tsai, Y.T.; Cao, Y.; Chen, J.R.; Shu, C.M. Incompatible reaction for (3-4-epoxycyclohexane) methyl-3’-4’-epoxycyclohexyl-carboxylate (EEC) by calorimetric technology and theoretical kinetic mode. J. Therm. Anal. Calorim. 2014, 116, 1445–1452. [CrossRef] 7. Wang, C.; Yang, Y.; Tsai, Y.T.; Deng, J.; Shu, C.M. Spontaneous combustion in six types of coal by using the simultaneous thermal analysis-Fourier transform infrared spectroscopy technique. J. Therm. Anal. Calorim. 2016, 126, 1591–1602. [CrossRef] 8. Tsai, Y.T.; Lin, S.Y.; Tong, J.W.; Chen, W.C.; Chen, W.T.; Shu, C.M. Incompatible hazard investigation of a cycloaliphatic epoxy resin using green analytical method. J. Therm. Anal. Calorim. 2015, 122, 1135–1141. [CrossRef] 9. Wanasekara, N.; Chalivendra, V.; Calverta, P. Sub-micron scale mechanical properties of polypropylene ﬁbers exposed to ultraviolet and thermal degradation. Polym. Degrad. Stab. 2011, 96, 432–437. [CrossRef] 10. Wang, L.X.; Wang, X.G. Progress of the trans-cis isomerization mechanism of azobenzene. Chemistry 2008, 71, 243–248. [CrossRef] 11. Wang, L.X.; Tuo, X.L.; Xu, J.; Li, S.N.; Wang, X.G. Effect of trans-cis isomerization on the sensitivity of energetic azo-compounds. Acta Phys. Chim. Sin. 2008, 24, 1756–1760. [CrossRef] 12. Yang, B.; Ahotupa, M.; Määttä, P.; Kallio, H. Composition and antioxidative activities of supercritical

CO2-extracted oils from seeds and soft parts of northern berries. Food Res. Int. 2011, 44, 2009–2011. [CrossRef] 13. Albrektiene,˙ R.; Rimeika, M.; Zalieckiene,˙ E. Determination of organic matter by UV absorption in the ground Water. J. Environ. Eng. Landsc. 2012, 20, 163–167. [CrossRef] 14. Lu, K.M.; Lee, W.J.; Chen, W.H.; Lin, T.C. Thermogravimetric analysis and kinetics of co-pyrolysis of raw/torreﬁed wood and coal blends. Appl. Energy 2013, 105, 57–65. [CrossRef] 15. Malow, M.; Wehrstedt, K.D. Prediction of the self-accelerating decomposition temperature (SADT) for liquid organic peroxides from differential scanning calorimetry (DSC) measurements. J. Hazard. Mater. 2005, 120, 21–24. [CrossRef][PubMed] 16. Tsai, Y.T.; You, M.L.; Qian, X.M.; Shu, C.M. Calorimetric techniques combined with various thermokinetic models to evaluate incompatible hazard of tert-butyl peroxy-2-ethyl hexanoate mixed with metal ions. Ind. Eng. Chem. Res. 2013, 52, 8206–8215. [CrossRef] 17. Tsai, L.C.; Tsai, Y.T.; Lin, C.P.; Liu, S.L.; Wu, T.C.; Shu, C.M. Isothermal versus non-isothermal calorimetric technique to evaluate thermokinetic parameters and thermal hazard of tert-butyl peroxy-2-ethyl hexanoate. J. Therm. Anal. Calorim. 2012, 109, 1291–1296. [CrossRef] 18. Feng, B.; Yu, M.; Ji, J.; Liu, L.; Ben, M.; Sun, J. Thermal properties and thermal degradation kinetics of polysulfonamide based single polymer composites. Polym. Mater. Sci. Eng. 2016, 32, 42–47. [CrossRef] Molecules 2017, 22, 2219 9 of 9

19. Vyazovkin, S.; Burnham, A.; Criado, J.; Pérez-Maqueda, L.; Popescu, C.; Sbirrazzuoli, N. ICTAC Kinetics Committee recommendations for performing kinetic computations on thermal analysis data. Thermochim. Acta 2011, 520, 1–19. [CrossRef] 20. Hou, H.Y.; Liao, T.S.; Duh, Y.S.; Shu, C.M. Thermal hazard studies for dicumyl peroxide by DSC and TAM. J. Therm. Anal. Calorim. 2006, 1, 167–171. [CrossRef] 21. Liu, S.H.; Shu, C.M. Advanced technology of thermal decomposition for AMBN and ABVN by DSC and VSP2. J. Therm. Anal. Calorim. 2015, 121, 1–8. [CrossRef] 22. Omrani, A.; Simon, L.C.; Rostam, A.A.; Ghaemy, M. Cure kinetics, dynamic mechanical and morphological properties of epoxy resin–Im6NiBr2 system. Eur. Polym. J. 2008, 44, 769–779. [CrossRef] 23. Doyle, C.D. Estimating Thermal stability of experimental polymers by empirical thermogravimetric analysis. Anal. Chem. 1961, 22, 77–79. [CrossRef] 24. Shu, C.M.; Chang, Y.H.; Chiu, C.W. Evaluation on the thermal stability and hazards behaviors of advn using green thermal analysis approach. J. Civ. Eng. Archit. 2016, 10, 280–290. [CrossRef] 25. Málek, J. The applicability of Johnson-Mehl-Avrami model in the thermal analysis of the crystallization kinetics of glasses. Thermochim. Acta 1995, 267, 61–73. [CrossRef] 26. Tseng, J.M.; You, M.Y.; Chen, S.L.; Hwang, W.T.; Gupta, J.P.; Shu, C.M. Runaway effects of nitric acid on methyl ethyl ketone peroxide by TAM III tests. J. Therm. Anal. Calorim. 2009, 96, 789–793. [CrossRef] 27. Egerton, T.A. UV-absorption-the primary process in photocatalysis and some practical consequences. Molecules 2014, 19, 18192–18214. [CrossRef][PubMed] 28. Lin, C.P.; Tseng, J.M. Green technology for improving process manufacturing design and storage management of organic peroxide. Chem. Eng. J. 2012, 180, 284–292. [CrossRef] 29. Matthes, P.R.; Schönfeld, F.; Zottnick, S.H.; Müller-Buschbaum, K. Post-synthetic shaping of porosity and crystal structure of Ln-Bipy-MOFs by thermal treatment. Molecules 2015, 20, 12125–12153. [CrossRef] [PubMed]

Sample Availability: Samples of the ADVN are available from the authors.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).