Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives

Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Paulo Bruno Pontes Serra Mestrado em Química Departamento de Química e Bioquímica 2012/2013 Supervisor Luís Manuel das Neves Belchior Faia dos Santos Professor associado, Faculdade de Ciências da Universidade do Porto. I FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives II Equation Chapter 1 Section 1 III FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives IV Acknowledgements I would like to thank all the support provided by Professor Fulem and Professor Růžička, along my internship in ICT Prague. If I had a successful stay, it was due to their support along the eight months lived in Prague. I would also like to thank all my friends that I met in Prague. They provided a great experience that I will carry with me for the rest of my life. Agradecimentos Gostaria de agradecer a todas as pessoas que trabalharam comigo de perto e que me ajudaram a criar este trabalho. Gostaria de realçar, nomeadamente, o Professor Luís Santos pela sua dedicação e apoio prestado, também à Marisa Rocha por toda a sua disponibilidade e apoio prestado e ao Filipe Ribeiro por todos os seus conselhos. Também gostaria de agradecer aos meus amigos da Faculdade de Ciências da Universidade do Porto que me acompanharam e apoiaram durante todos estes anos de estudo. Por fim, gostaria de agradecer à minha família e restantes amigos que me ajudaram a tornar no homem que sou hoje. V FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives VI Resumo Este trabalho foca o estudo térmico, a medição de capacidades caloríficas em fase condensada e o estudo de transições de fase de líquido iónicos. Foram estudadas dua famílias de líquidos iónicos, uma baseada no catião 1-benzil-3-metilimidazólio conjugado com quatro aniões (tetrafluoroborato; hexafluorofosfato; 1,1,2,2-tetrafluoroetanossulfonato; bis(trifluorometilsulfonil)imida) e a série 1-alquil-3-metilimidazólio: metil, -etil, -propil, -butil, - pentil, -hexil. –heptil, -octil, -nonil, -decanil and -dodecanil-3-metilimidazólio com o anião hexafluorofosfato. O efeito de diferentes aniões na presença de um catião comum foi estudado na série do 1-benzil-3-metilimidazólio, assim como o impacto do comprimento da cadeia alquílica no catião da série do 1-alquil-3-metilimidazólio. O estudo térmico, cristalinidade, temperatura de vitrificação, estudo de transições de fase sólido-sólido e fusão, foram investigadas por calorimetria diferencial de varrimento na gama de temperaturas de 183 K a 423 K. Com base nos resultados obtidos foram derivadas as entalpias e entropias das transições de fase mais relevantes. As capacidades caloríficas das fases condensadas foram medidas usando o método contínuo e de salto de temperatura no intervalo de temperaturas de 253 K a 353 K por microcalorimetria diferencial de varrimento. Os resultados obtidos para temperaturas de transições de fase e capacidades caloríficas foram comparados com valores da literatura e com métodos de estimativa de contribuição de grupos. Adicionalmente, é ainda apresentada uma análise comparativa do efeito do anião nos líquidos iónicos derivados do 1-benzil-3-metilimidazólio, assim como o efeito do grupo benzil. O efeito do tamanho da cadeia alquílica é explorado nas propriedades da série 1-alquil-3-metilimidazólio hexafluorofosfato. VII FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives VIII Abstract This work is focused on the thermal study, condensed phases heat capacities measurements and phase behaviour studies of ionic liquids. It was studied two families, one based on the 1-benzyl-3-methylimidazolium cation conjugated with 4 anions (tetrafluoroborate; hexafluorophosphate; 1,1,2,2-tetrafluoroethanesulfonate; bis(trifluoromethylsulfonyl)imide), and the 1-alkyl-3-methylimidazolium series: -methyl; -ethyl; -propyl; -butyl; -pentyl; -hexyl; –heptyl; - octyl; -nonyl; -decyl and -dodecyl-3-methylimidazolium with the anion hexafluorophosphate. The effect of the anion considering a common cation was explored for the 1-benzyl-3methylimidazolium based ionic liquids, and additionally, it was evaluated the impact of the alkyl side chain in the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquid series. The thermal study, crystallinity, glass transition temperature, solid-solid phase transitions and melting studies, were investigated by the differential scanning calorimeter in the temperature range from 183 K to 423 K. With the obtained results, the most relevant transitions enthalpy and entropy were derived. The heat capacities of the condensed phases were measured by continuous and step method in the temperature interval from 253 K to 353 K by differential scanning microcalorimeter. The obtained results for phase transitions temperatures and heat capacities were compared with available literature values as well as with group contribution estimation methods. A comparative analysis of the anion effect in the 1-benzyl-3-methylimidazolium ionic liquids, and the benzyl group effect is also presented. The alkyl side chain length effect on the properties of 1-alkyl-3-methylimidazolium series is explored. XV FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Figure 1.14 – Example of different crystallization degrees in ILs, performed for the same sample, under the same experimental conditions, in the temperature range of 183 to 423 K, at 5 K∙min-1. a, and b are the glass transition and the cold respectively............................................27 Figure 2.1 - Scheme of the drying system: 1Sample vessel; 2Cold trap (glass); 3Cooling system; 4Valves; 5Vacuum pump system. ...............................................................36 Figure 2.2 - Components of the vacuum system: aEdwards RV3 rotary vacuum pump; bPfeiffer Balzers TCP 310 turbo pump controller; cPfeiffer Balzers TPU-170 turbomolecular vacuum pump. ..........................................................................................................................37 Figure 3.1 - Example of a sample data analysis using TA Universal Analysis software. 41 Figure 3.2 - Example of the 3 measurements performed for [C2C1im][PF6] phase transition analysis. ....................................................................................................................42 Figure 3.3 - Thermal analysis for [Bnmim][BF4] performed from 183 to 423 K at a 5 K∙min-1 rate. ..............................................................................................................................43 Figure 3.4 - Thermal analysis for [Bnmim][PF6] performed from 183 to 348 K at a 5 K∙min-1 rate. ..............................................................................................................................44 Figure 3.5 - Thermal analysis for [Bnmim][NTf2] performed from 183 to 348 K at a 5 K∙min-1 rate. ..............................................................................................................................44 Figure 3.6 - Thermal analysis for [Bnmim][ C2F4HSO3] performed from 183 to 348 K at a 5 K∙min-1 rate. ...........................................................................................................................45 Figure 3.7 - Thermal analysis for [C2C1im][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................46 Figure 3.8 - Thermal analysis for [C3C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................47 Figure 3.9 - Thermal analysis for [C4C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................47 Figure 3.10 - Thermal analysis for [C5C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................48 Figure 3.11 - Thermal analysis for [C6C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................49 Figure 3.12 - Thermal analysis for [C7C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................49 Figure 3.13 - Thermal analysis for [C9C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................50 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives XVI Figure 3.14 - Thermal analysis for [C10C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................51 Figure 3.15 - Thermal analysis for [C12C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. ..............................................................................................................................51 Figure 3.16 – Graphical representation of the Tg, Ts-s, Tcc, and Tm as a function of the number of carbons on the alkyl side chain of the [CnC1im]+ cation. ............................................53 Figure 3.17 - Enthalpy dependence of the cation alkyl chain size, for melting. ...............55 Figure 3.18 - Entropic dependence of the cation alkyl chain size, for melting. ...............55 Figure 3.19 - Relative deviation of the experimental heat capacities Cpexp for [Bnmim][PF6] from the smoothed values Cplf. The black line represents the continuous method values and the squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). ................................................................58 Figure 3.20 - Relative deviation of the experimental heat capacities Cpexp for [Bnmim][PF6] not dried, from the smoothed values Cplf. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). ................................................................60 Figure 3.21 - Relative deviation of the experimental heat capacities, Cpexp, for [Bnmim][PF6] from the fit values, Cplf, for both dried and not dried samples. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). ..................................60 Figure 3.22 - Relative deviation of the experimental heat capacities Cpexp for [Bnmim][BF4] from the smoothed values Cplf. The black line represents the continuous method values and the squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). ................................................................61 Figure 3.23 - Relative deviation of the experimental molar heat capacities Cpexp for [Bnmim][C2F4HSO3] for step method. The squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). ............................63 Figure 3.24 - Relative deviation of the experimental molar heat capacities Cpexp for [Bnmim][C2F4HSO3] for step method. The squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). ............................65 Figure 3.25 – Molar heat capacity dependence with the temperature, for [Bnmim][PF6], [Bnmim][NTf2], [Bnmim][BF4], and [Bnmim][C2F4HSO3]. All compounds were studied under the same experimental conditions. ..................................................................................................67 XVII FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Figure 3.26 - Heat capacity dependence with temperature of [Bnmim][PF6], [Bnmim][NTf2], [Bnmim][BF4], and [Bnmim][C2F4HSO3]. The dashed lines are extrapolations made for each physical state (solid and liquid). .........................................................................68 Figure 3.27 - Apparent molar heat capacities, at 298 K, as temperature function for the [CnC1im] studied, where n = 2 - 10 and 12. ................................................................................70 Figure 3.28 – Molar heat capacities as temperature function, at 298.15 K, for the [CnC1im][PF6] studied, where n = 2 - 10 and 12 (empty symbols are for solid phase and full symbols are for liquid phase). ...................................................................................................70 Figure 3.29 - Temperature dependence on [CnC1im][PF6] (where n = 2-10 and 12) the molar heat capacities, at 355 K, in the liquid phase as function of the side alkyl chain. .............71 Figure 3.30 - Molar heat capacities, at T = 298.15 K, as function of the number of carbon atoms in the alkyl side chain of the cation, n(C), for [CnC1im][PF6] (with n = 2 – 8, 10, 12) and [CnC1im][NTf2] (with n = 2 – 8, 10, 12). ......................................................................................72 Figure 3.31 - Specific heat capacities, at T = 298.15 K, as function of the number of carbon atoms in the alkyl side chain of the cation, n(C), for [CnC1im][PF6] (with n = 2 – 8, 10, 12) and [CnC1im][NTf2] (with n = 2 – 8, 10, 12). ...............................................................................73 Figure 3.32Volumic heat capacities at 298 K, as function of the number of carbon atoms in the alkyl side chain of the cation .................................................................................74 Figure 3.33 - Comparison between the data obtained for [CnC1im][PF6], where n = 2 - 10 and 12, and the data from group contribution method and literature data found for the same compounds. ..............................................................................................................................78 Figure 3.34 – Relative deviation from GCM values (0 line), of the experimental and literature values for [CnC1im][PF6], where n = 2 - 10 and 12. “0 line” stands for the values obtained by linear fit. .................................................................................................................78 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives XVIII Table index Table 2.1 – Summary of the 1-Benzyl-3-methylimidazolium [Bnmim]+, based ionic liquids. ......................................................................................................................................34 Table 2.2 - Summary of the 1-alkyl-3-methylimidazolium hexafluorophosphate, [CnC1im][PF6], ionic liquids series ..............................................................................................35 Table 2.3 - Initial mass (before drying), final mass (after drying) and mass loss for the studied ionic liquids. ..................................................................................................................37 Table 3.1 - Phase transition temperatures: Tg, Ts-s, Tcc and Tm obtained using the TA Instruments Q1000 DSC. ..........................................................................................................52 Table 3.2 – Melting temperature, entalpies and entropies of melting of the studied ILs. 54 Table 3.3 – Quadratic fit parameters for all the studied compounds, determined from step method measurements. .....................................................................................................56 Table 3.4 - Linear fit parameters for all the studied compounds, determined from step method values...........................................................................................................................57 Table 3.5 - Experimental molar heat capacities (J∙K-1∙mol-1) for solid [Bnmim][PF6] (dried and not dried sample) and the data derived from quadratic fitting for both step and continuous method. .....................................................................................................................................59 Table 3.6 - Experimental molar heat capacities (J∙K-1∙mol-1) for solid and liquid phases of [Bnmim][BF4] and results derived from fitted equation for both step and continuous method. ....62 Table 3.7 - Experimental molar heat capacities (J∙K-1∙mol-1) for solid and liquid phases of [Bnmim][C2F4HSO3] and results derived from fitted equation for both step and continuous method. .....................................................................................................................................64 Table 3.8 - Experimental molar heat capacities (J∙K-1∙mol-1) for liquid [Bnmim][NTf2] and results derived from fitted equation for both step and continuous method. ................................66 Table 3.9 – Solid and liquid molar heat capacities of the [Bnmim] series at 298.15 K and respective molar weight. ...........................................................................................................67 Table 3.10 – Molar heat capacities for the 1-alkyl-3-methylimidazolium serie at 298.15 K and respective molar weight. .....................................................................................................69 Table 3.11 - Tg and Tm experimental results obtained from [Bnmim][BF4], [Bnmim][PF6], [Bnmim][C2F4HSO3], [Bnmim][NTf2] and [CnC1im][PF6], where n = 2 - 10 and 12. And XIX FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives comparison with the estimated values computed from Lazzús group contribution method [34, 43] and literature data from Chun et al [44]. .............................................................................75 Table 3.12 - Tg and Tm experimentally obtained for [Bnmim][BF4], [Bnmim][PF6], [Bnmim][C2F4HSO3], [Bnmim][NTf2] and the estimated values from Gharagheizi et al [46, 47] GCM. ........................................................................................................................................76 Table 3.13 - Experimental molar heat capacity data, at 298 K, obtained for [CnC1im][PF6], where n = 2 - 10 and 12, and comparison with Gardas et al [27] group contribution method and literature data from Paulechka [45]. ...........................................................................................77 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives XX Abbreviations list ILs - Ionic Liquids DSC - Differential Scanning Calorimetry Tg - Glass transition temperature Ts-s – Solid-solid transition temperature Tm - Melting temperature Tcc – Cold crystallization temperature cp – Specific heat capacity at constant pressure Cp,m – Molar heat capacity at constant pressure Cplf – Heat capacity at constant pressure from linear fit Cpexp – Heat capacity at constant pressure experimental GCM – group contribution method RTILs – Room temperature ionic liquids MALDI - Matrix-assisted laser desorption/ionization TOF - Time-of-flight mass spectrometer GC – Gas chromatography ΔH – Enthalpy variation ΔT – Temperature variation T - Temperature ΔS – Entropy variation ΔG – Gibbs energy variation XXI FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives mW – Milliwatt CAS number – Chemical abstracts service number dev. – deviation eq. – Equation J – Joule g - gram K – Kelvin n [C] – number of Carbons in the alkyl chain s – second exp – experimental lit – literature 1 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1. Introduction FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 2 1.1. General introduction This thesis is based on the work performed in the Faculty Vysoká Škola ChemickoTechnologická v Praze, VSCHT, in Prague. Along the 8 months of the Erasmus placement, the phase transitions and heat capacities of several ionic liquids were studied with the collaboration of Professor Michal Fulem and Professor Květoslav Růžička. 1.2. Motivation and Aim of the Work The present work is focused on the thermodynamics of phase transitions and heat capacities of the condensed phases of some ionic liquids. The studied samples are based on imidazolium cation. This organic cation features some interesting properties that lead to its family study [1]:  thermal and electrochemical stability;  wide liquidus range;  low reactivity with water;  low viscosity;  solvent properties adjustment. Two variations of this cation were chosen, allowing the work division in two parts and to study different properties. The first group possesses a bulky variation of the cation: the 1Benzyl-3-methylimidazolium (Figure 1.1) with each nitrogen connected with distinct groups. In one, a methyl group is connected (small size) and in the other a benzyl group is bonded (a bulky group) as seen in Figure 1.1. Those features lead to a delocalization in the cation charge and a big asymmetry. This cation was tested with 4 different anions that are commonly used nowadays: two small sized and symmetric inorganic anions, the BF4 and PF6; and two big sized 9 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.5. Differential Scanning calorimetry, DSC The Differential Scanning calorimetry, or DSC, is a calorimeter where the sample is submitted to a controlled temperature program and its behaviour is recorded for phase transition analysis and heat capacity calculation [7]. For the measurement, the sample is inserted in a metallic pan (usually aluminium or platinum, to allow a fast heat transfer between the sample and its surroundings) and well closed to avoid mass loss along measurements. Along the measurements, the results are affected by the pan heat retention and to remove this effect, an empty reference pan is used along the sample measurement. This reference pan can be in the same furnace as the sample or in a separate one, distinguishing the two different DSCs:  Power compensation DSC – sample and reference are separated and the power measurement is done by comparison between difference of energy supply to the sample and reference cavities;  Heat Flow DSC – sample and reference share the same furnace and the power measurement is based in the incoming/outcoming - heat flow difference between the sample and the reference cell. The DSC’s are commonly used in the development, quality and process control in many industrial fields and for thermal characterization of construction materials, polymeric materials, composites, chemicals and pharmaceuticals etc. [21]. DSC can be used as a tool for:  Thermal behaviour;  Phase transitions analysis;  Heat capacity measurements;  Process Enthalpy and Entropy evaluations;  Studies about the heat generation in biological systems. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 10 In the studies of ILs, two different DSC’s were used: a TA Instruments Q1000 DSC (used in phase transitions analysis) and a Setaram μDSCIIIa (for the heat capacity determination). 11 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.6. The Differential Scanning Calorimeters In the ILs studies, two different DSC were used: a TA Instruments Q1000 DSC (for phase transitions analysis, Figure 1.3) and a Setaram μDSCIII (for heat capacity determinations, Figure 1.5). 1.6.1. TA Instruments Q1000 Figure 1.3 - The TA Instruments Q1000 used in this work for phase transitions measurements. The measurements with the TA Instruments Q1000 were performed using a continuous method, using a linear heating rate of 3 K•min-1, from 183 K to 403 K. Two isothermal delays of 1800 s were used in the beginning and at the end of the measurement to allow the DSC signal and sample stabilization (ionic liquids are deeply affected by their thermal history which depending on the heating/cooling rate induces different types of crystallization, so to avoid the study of different crystal phases the sample is put under a high cooling rate and stabilized before beginning the heating for thermal analysis). The pans used in the TA Instruments Q1000 are aluminum made, and are depicted in Figure 1.4 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 12 Figure 1.4 – The 10 mm3 Tzero series hermetic pans used in the TA Instruments Q1000. This calorimeter used a technical nitrogen gas flow (from SIAD Czech Company) with a guaranteed purity (from the producer) of more than 99.99% (the gas flow used was around 50 mL•min-1). Other characteristics from this heat flow DSC are listed below [22]:  Temperature range from 183 K673 K;  Temperature Accuracy of 0.1 K;  Temperature Precision of 0.05 K;  Calorimetric Precision (metal standards) of 1%;  Power sensitivity of 0.2 W;  50-position autosampler;  Digital mass flow controller. 13 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.6.2. Setaram μDSC IIIa Figure 1.5 - The Setaram μDSC IIIa used in this work for heat capacities measurements. The liquid heat capacity was measured with a Setaram DSC IIIa, in the range from 273 to 355 K and used the incremental temperature mode for the step method (described in chapter 1.7.2). Each 5 K step included a heating rate of 0.3 K•min-1 was used, between two isothermal delays of 2600 s. For continuous method, a heating rate of 0.3 K∙min-1 was used, between two isothermal delays of 2600 s. Obtained data was integrated using the Setaram software package SetSoft 2000. The typical mass of samples was 0.4 g to 1 g. The combined expanded uncertainty of the ICT heat capacity measurements is estimated to be Uc(Cp,m) = 0.01 Cp,m. [23], the measuring procedure was described in detail in Fulem et al [24]. In Figure 1.6 are presented the closed Hastelloy C (majorly made by Nickel, Chromium and Molybdenum) vessels, with volume of 1 cm3, used in the Setaram μDSC IIIa. Figure 1.6 - The μDSC vessels used in the Setaram μDSC IIIa. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 14 The temperature of the calorimeter is regulated by means of a cooled water bath. The Setaram μDSC IIIa presents characteristics such as [22]:  Temperature range from 253 - 393 K;  Scanning rate from 0.001 K•min-1 to 1.2 K•min-1 in the whole temperature range;  Detection limit from 0.2 until 2 µW;  Resolution of 40 nW;  Two scanning modes - Isothermal and differential. 15 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.7. Heat capacity The enthalpy change of a sample with the temperature can be related with the heat capacity of the sample at constant pressure, Cp. The heat capacity at constant pressure, Cp, is given by the following equation [25]: () pp H CT T      (1.1) The heat capacity at constant pressure is used to relate the enthalpy change with the temperature. For infinitesimal changes of temperature, at constant pressure: dd p H C T (1.1) For a short temperature interval the heat capacity could be taken as constant and the enthalpy of a process associated with a the temperature change can be derived from equation 1.3. p H C T   (1.3) And at constant pressure, related with the heat involved in the process, equation 1.4. pp q C T  (1.4) The heat capacity can be derived from the balance of the heat transferred to the sample, at constant pressure along the temperature change. The heat capacity is one of the basic thermophysical and thermodynamic properties that characterizes a compound. In thermochemistry, heat capacity is the amount of energy as heat required to raise the temperature of a sample by one degree [26]. The isobaric heat capacity is required for the calculation of temperature dependence of fundamental thermodynamic functions (some of them were computed in this work as the case of the enthalpy and entropy change). The knowledge of those thermodynamic functions in ILs are important and essential for the understanding of the physical-chemistry properties and behavior of the ILs [27]. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 16 1.7.1. Heat capacity measurements In all measurements, a typical methodology was used. This method consists in the combination of three different measurements to achieve the heat capacity of the sample (the reference cell is measured simultaneously): 1. An empty sample cell measurement (the obtained peak area is AB); 2. A measurement where the sample cell is filled with the reference material, in this case sapphire (the obtained peak area Asapp), with known heat capacity (C(p,sapp)); 3. And a final measurement where the sample cell is filled with the sample (the obtained peak area - As). A graphical representation can be seen in Figure 1.7. Figure 1.7 - Graphical representation of the three-step method used in the measurements. Most of the images presented in the figures were taken from print shots of the data analysis software and the temperature labelling is presented in Celsius (ºC). Along the thesis text the temperature, T is presented and discussed in Kelvin (K). The experimental temperature data was converted to Kelvins (K) by the following equation: ( ) 273.15 ( )T K T C   (1.5) 17 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives The experimental heat capacity measurements were done following two different methodologies: step method and continuous method:  Step method - the temperature of the sample is programmed in step mode with 5 K “jumps” at 0.3 K∙min-1 (before and after the temperature jump, the temperature keep in isothermal mode for 2600 s);  Continuous method - the temperature of the sample is programmed to change continuous with a fixed temperature scanning rate of 0.3 K∙min-1 along all the measuring temperature interval. Figure 1.8 and Figure 1.9 presents a schematic representation of the two methodologies: step method and continuous method, respectively. Figure 1.8 – Schematic representation of the temperature program and heat flow in the step method (each temperature jump is 5 K at 0.3 K∙min-1, between two temperature stabilizations of 2600 s). 0 3 6 9 0 2 0 3 6 9 260 280 300 Heat flow (mW) Time (h) Time (h) T/K FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 18 Figure 1.9 – Schematic representation of the temperature program and heat flow in the continuous method (0.3 K∙min-1). 0 3 6 0 2 0 3 6 240 300 360 Heat flow (mW) Time (h) Time (h) T/K 25 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.8.2. Crystallization Crystallization occurs when there is a formation of solid crystals, from a primary liquid phase (when cooling the sample), through two processes:  Nucleation;  Nuclei growth. Other form of crystallization occurs when the sample is slowly heated above the Tg, forming crystalline structures – the cold crystallization, Tcc [29]. Both processes occur from a less organized structure, amorphous liquid or solid, to a more organized or crystalline structure. An amorphous solid state is characterized by a not totally crystallized compound where some molecules have freedom to move, explaining the rubbery state that characterizes this phase. This is presented in the thermogram as multiple melting or crystallization peaks, what means that such samples are difficult to crystallize in one step and, in a general way, those compounds possess long alkyl side chains that tend to crystallize in steps, due to C-C (carbon-carbon) bond rotation. These solid-solid phase transitions occurring before the melting point are often referred to as rotator phases, which generally possess plastic properties [30]. These need even longer cooling so that one-step crystallization can be reached as seen in Figure 1.13. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 26 Figure 1.13 - In this example, different peaks can be seen, along the sample heating. All were performed under the same experimental conditions, in the temperature range of 183 to 423 K, at 5 K∙min-1. a, b and c are the glass transition, the cold crystallization and the melting peak, respectively. Those different peaks size are due to different crystallization degrees, this is proved by the fact of not having a cold crystallization peak in the first measurement and different peak sizes in the following measurements. In the first measurement, since the compound is totally crystalline, the heat flow is higher and there is no cold crystallization peak. In the second measurement, due to a fast cooling, some of the sample didn’t have time to crystallize and so there is a lower heat flow signal (so, a lower heat capacity) and a first, smaller cold crystallization peak, from the amorphous crystal part. In the last measurement, the sample is totally in an amorphous state and so, the heat flow is represented by the lower heat flow signal and the higher crystallization peak. -100 -50 0 50 100 150 Temperature (°C) -3 -2 -1 0 Heat Flow (W/g) -90 -40 10 60 110 Temperature (°C) Sample: AI250_[Bnmim]PF6 Size: 4.5800 mg Method: Standard_IL_-90_65_5 Comment: AI250_[Bnmim]PF6 DSC File: I:...\IonicLiquids\AI250_[Bnmim]PF6.001 Operator: KRu Run Date: 2013-03-28 15:57 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments b c c a 27 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives The use of a fast cooling rate can show amorphous areas in samples and slow cooling leads to higher degrees of crystallization. This can be better seen in Figure 1.14. Figure 1.14 – Example of different crystallization degrees in ILs, performed for the same sample, under the same experimental conditions, in the temperature range of 183 to 423 K, at 5 K∙min-1. a, and b are the glass transition and the cold respectively. -100 -50 0 50 100 150 Temperature (°C) -1.5 -1.2 -0.9 -0.6 -0.3 Heat Flow (W/g) -60 -40 -20 0 20 40 Temperature (°C) Sample: AI250_[Bnmim]PF6 Size: 4.5800 mg Method: Standard_IL_-90_65_5 Comment: AI250_[Bnmim]PF6 DSC File: I:...\IonicLiquids\AI250_[Bnmim]PF6.001 Operator: KRu Run Date: 2013-03-28 15:57 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments b c a FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 28 1.8.3. Glass transition A glass transition or the liquid-glass transition, Tg, is the reversible transition in amorphous materials from a molten or rubberlike state into a hard and relatively brittle state and vice-versa [31]. This is an important characteristic, considering that some ILs only show a crystallization peak – cold crystallization - after heating from low temperatures. Usually, the cold crystallization peak can be seen from 20 to 30 temperature degrees after the glass transition. Also, some ILs do not present crystallization peak, until the glass transition temperature is achieved, staying in a supercooled liquid state [32]. In the previous cases, the transition kinetics, mainly governed by the cooling rate (rapid or slow cooling), plays an important role, so, in order to obtain reliable thermal phase behavior data, long equilibration times and small samples are needed. Usually, the Tg can be difficult to determine because it can take place over a wide temperature range and is highly dependent on conditions such as the measurement method and pressure [33]. For 1-alkyl3-methylimidazolium salts, Tg recorded are typically in the region between 203 K and 183 K [34]. The Tg can also be related with the sample melting temperature as shown in eq. 1.8 (temperature values are presented in K) [35]: Tg/Tm (K) = 2/3 (1.8) And for most of the studied ILs: Tg/Tm (K) = [2/3; 3/4] (1.9) 29 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.8.4. Melting point Melting point is the temperature which, at atmospheric pressure, there is a phase transition from solid to liquid state. Typically, at this temperature, the sample is in solid-liquid equilibrium and remains isothermal, until the whole sample is melted.[36] In the ILs, the melting point is deeply influenced by many factors [37]:  different cation and anion size can lead to a lower fusion temperature (weaker electrostatic interactions in the crystal lattice);  symmetrical ions leads to higher fusion temperatures (better crystal packing);  great charge delocalization leading to lower fusion temperature. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 30 1.9. Group contribution methods Nowadays, the number of known ILs is around thousands (and their number is still growing!) and so, to analyse all of them not feasible. In addition, some properties such as Tg, cannot be measured due to apparatus limitations regarding for example working temperature range. To solve this kind of problems, the prediction methods are developed and used to obtain missing data. The most used prediction methods are those based on a group contribution concept, where the group is defined as a set of atoms with a well-defined value for a given property. From the established contributions, one can calculate the physicochemical properties of a compound of interest (if all contributions, to which the compound is divided, are available). For example, the glass transition temperature can be estimated from the following groups sum (eq.1.10) [34]: (1.10) Where all the groups contribute for the first equation portion plus some corrections necessary, previously determined through studies (as well as all the values defined for each group and respective corrections). g T contribution of composing groups corrections where necessary    31 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 1.9.1. Group contribution method for heat capacities For heat capacities an estimation method used by Gardas et al [27] is presented in equation 1.11.: 2 ( / ) 100 100 pTT C T K R A B D                    (1.11) where R is the gas constant (R=8.314462 J∙K-1∙mol-1), T is the absolute temperature in K, and A, B and D are constants obtained from the following relations (eq: 1.12; 1.13 and 1.14): 1 k ii i A na   (1.12) 1 k ii i B nb   (1.13) 1 k ii i D nd   (1.14) Where n is the number of times that the i group appears, and a, b and d are fitted parameters [27]. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 32 1.9.2. Group contribution method for phase transitions For ionic liquids, phase transition temperatures can be predicted through the equation 1.15, due to the effect of the pair anion/cation as suggested by Lazzús [34]: g i ci j aj T C n t n t     , (1.15) where Tg (K) is the temperature of glass transition, C is a constant, ni and nj are the number of times that the groups i and j appear in the compound, Δtci is the cation contribution and Δtaj is the anion contribution for the molecule. An identical equation was used for melting temperatures in the GCM. 33 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 2. Experimental part Equation Section 2 Equation Chapter (Next) Section 1 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 34 2.1. Ionic liquids description Some of the ionic liquids under study are viscous liquids (transparent to yellow colour), other are solids (white powder) with molecular weights between 250 and 450 g mol-1 (the relative atomic masses used were those recommended by the IUPAC Commission in 2007 [38]). All of the ionic liquids were purchased from IOLITEC with a stated purity of better than 99%, which is the necessary for this kind of studies. 2.1.1. 1-Benzyl-3-methylimidazolium ([Bnmim]+) based ILs The first group consists of four ILs (Table 2.1) with a common cation – the 1-benzyl3-methylimidazolium, and four different anions which displays different characteristics between them (volume, charge dispersion, cation interaction): the BF4, the PF6 (these two are amply investigated nowadays), the NTf2 and the C2F4HSO3 (these last two are recent and more is needed to discover about them. The information regarding this ionic liquid family is compiled in Table 2.1. Table 2.1 – Summary of the 1-Benzyl-3-methylimidazolium [Bnmim]+, based ionic liquids. Compound name Abbreviation CAS number Supplier/ Purity Molecular weight (g/mol) Compound visual aspect 1-Benzyl-3-methylimidazolium tetrafluoroborate [Bnmim] [BF4] 500996-04-3 IoLiTec/ >99% 260.04 White powder 1-Benzyl-3-methylimidazolium hexafluorophosphate [Bnmim] [PF6] 433337-11-2 IoLiTec/ >99% 318.20 White powder 1-Benzyl-3-methylimidazolium 1,1,2,2-tetrafluoroethanesulfonate [Bnmim] [C2F4HSO3] n/a IoLiTec/ >99% 354.32 Amorphous white solid 1-Benzyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide [Bnmim] [NTf2] 433337-24-7 IoLiTec/ >99% 453.38 Yellow viscous liquid 41 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.1. Phase transitions The phase behavior was investigated from 183 to 423 K with a differential scanning calorimeter (TA Q1000, TA Instruments, USA) using the continuous method with a heating rate of 5 K∙min-1. Samples of about 0.1 g of each ionic liquid were placed in the crucibles and weighted by an analytical balance with a readability of 0.01 mg. After the measurements, the obtained data was analyzed using the TA Universal Analysis software as can be observed in Figure 3.1 and Figure 3.2. Figure 3.1 - Example of a sample data analysis using TA Universal Analysis software. 53.42°C 58.12J/g 52.37°C 59.15J/g 52.32°C 57.22J/g -100 -50 0 50 100 150 Temperature (°C) -0.8 -0.6 -0.4 -0.2 0.0 0.2 Heat Flow (W/g) -100 -50 0 50 100 150 Temperature (°C) Sample: AI248_C2C1im[PF6] Size: 10.8900 mg Method: Standard_IL_-90_150_5 Comment: AI248_C2C1im[PF6] DSC File: I:...\IonicLiquids\AI248_C2C1im[PF6].002 Operator: KRu Run Date: 2013-04-02 21:25 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 42 Figure 3.2 - Example of the 3 measurements performed for [C2C1im][PF6] phase transition analysis. 49.00min 47.22min 58.12J/g 142.67min 140.85min 59.15J/g 236.50min 234.65min 57.22J/g -100 -50 0 50 100 150 Temperature (°C) -0.8 -0.6 -0.4 -0.2 0.0 0.2 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI248_C2C1im[PF6] Size: 10.8900 mg Method: Standard_IL_-90_150_5 Comment: AI248_C2C1im[PF6] DSC File: I:...\IonicLiquids\AI248_C2C1im[PF6].002 Operator: KRu Run Date: 2013-04-02 21:25 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments 43 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.1.1. The 1-benzyl-3-methylimidazolium based ionic liquids The phase transitions results for each compound of the 1-benzyl-3methylimidazolium series are presented. The thermograms for each compound are presented in figures 3.3 to 3.6. [Bnmim][BF4] was solid at room temperature and liquid phase was reached at 336 K (62 ºC). Cold crystallization was not observed in the first measurement only appearing in the two following measurements at 282 K (45 ºC). Glass transition was detected at 235 K (-37 ºC). a is just an error occurred along the measuring. Figure 3.3 - Thermal analysis for [Bnmim][BF4] performed from 183 to 423 K at a 5 K∙min-1 rate. [Bnmim][PF6] was solid at room temperature and liquid phase was reached at 399 K (78 ºC). As observed for the previous ionic liquid, the cold crystallization only appeared in the two last measurements at 278 K (5 ºC). Glass transition was detected at 244 K (-28 ºC). 135.43min 9.29°C 44.93J/g -37.63°C 145.96min 63.29°C 63.95J/g 51.41min 63.39°C 64.31J/g 230.07min 9.46°C 45.22J/g -37.92°C 240.52min 63.31°C 64.22J/g -100 -50 0 50 100 150 Temperature (°C) -2.0 -1.5 -1.0 -0.5 0.0 0.5 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI263[Bnmim]BF4 Size: 13.5500 mg Method: Standard_IL_-90_150_5 Comment: AI263[Bnmim]BF4 DSC File: I:...\IonicLiquids\AI263[Bnmim]BF4.001 Operator: PS Run Date: 2013-04-02 02:12 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments a FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 44 Figure 3.4 - Thermal analysis for [Bnmim][PF6] performed from 183 to 348 K at a 5 K∙min-1 rate. The third compound is the [Bnmim][NTf2]. It was liquid at room temperature and solid phase was not reached in the temperature range studied, remaining in a subcooled liquid state until glass transition detection, at 216 K (-57 ºC). Figure 3.5 - Thermal analysis for [Bnmim][NTf2] performed from 183 to 348 K at a 5 K∙min-1 rate. 135.18min 5.83°C 47.88J/g -28.16°C 158.80min 126.46°C 76.10J/g -28.31°C 229.78min 6.31°C 48.97J/g 253.37min 126.61°C 79.31J/g 64.63min 131.17°C 69.73J/g -100 -50 0 50 100 150 Temperature (°C) -3 -2 -1 0 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI250_[Bnmim]PF6 Size: 4.5800 mg Method: Standard_IL_-90_65_5 Comment: AI250_[Bnmim]PF6 DSC File: I:...\IonicLiquids\AI250_[Bnmim]PF6.001 Operator: PS Run Date: 2013-03-28 15:57 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments -54.75°C -57.58°C -57.29°C -100 -50 0 50 Temperature (°C) -0.4 -0.3 -0.2 -0.1 0.0 0.1 Heat Flow (W/g) 050 100 150 200 Time (min) Sample: AI247_[Bnmim]NTf2 Size: 19.2600 mg Method: StandardIL_(-90_20m)_05_(130_10 Comment: AI247_[Bnmim]NTf2 DSC File: I:...\IonicLiquids\AI247_[Bnmim]NTf2.001 Operator: PS Run Date: 2013-03-28 02:39 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments 45 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives [Bnmim][C2F4HSO3] was also solid at room temperature and liquid phase was reached at 315 K (61 ºC). Cold crystallization was not reached in the first measurement appearing in the two measurements after. Glass transition was detected at 229 K (43 ºC), after the first measurement. Figure 3.6 - Thermal analysis for [Bnmim][ C2F4HSO3] performed from 183 to 348 K at a 5 K∙min-1 rate. 47.39min 43.10°C 64.31J/g 122.67min 42.30°C 63.80J/g 114.77min -6.18°C 46.68J/g 190.14min -6.02°C 43.19J/g -43.78°C 198.05min 42.52°C 58.42J/g -44.00°C -100 -50 0 50 Temperature (°C) -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 Heat Flow (W/g) 050 100 150 200 Time (min) Sample: AI246_[Bnmim]C2F4HSO3 Size: 9.2500 mg Method: StandardIL_(-90_20m)_05_(130_10 Comment: AI246_[Bnmim]C2F4HSO3 DSC File: I:...\AI246_[Bnmim]C2F4HSO3.001 Operator: PS Run Date: 2013-03-27 22:51 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 46 3.1.2. The 1-alkyl-3-methylimidazolium based ionic liquids The phase transitions results for each compound of the 1-alkyl-3-methylimidazolium series are presented, in the temperature range of 183 K to 423 K (-90 to 150 ºC), at a 5 K∙min-1 heating rate. The thermograms obtained for this ionicliquids family, [CnC1im][PF6], are presented in figures 3.7 to 3.15. [C2C1im][PF6] was solid at room temperature and the liquid phase was reached at 317 K (43 ºC). No glass transition or cold crystallization was detected, instead, a solid-solid transition at 232 K (-41 ºC) can be detected. Figure 3.7 - Thermal analysis for [C2C1im][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. [C3C1im][PF6] was also solid at room temperature and liquid phase was reached at 311 K (38 ºC). Neither the cold crystallization nor the glass transition were observed in the studied temperature range, but a solid-solid transition can be observed at 232 K (-41 ºC). 49.00min 47.21min 44.81°C 59.65J/g 142.67min 140.85min 43.58°C 58.21J/g 236.50min 234.65min 43.36°C 58.16J/g -100 -50 0 50 100 150 Temperature (°C) -0.8 -0.6 -0.4 -0.2 0.0 0.2 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI248_C2C1im[PF6] Size: 10.8900 mg Method: Standard_IL_-90_150_5 Comment: AI248_C2C1im[PF6] DSC File: I:...\IonicLiquids\AI248_C2C1im[PF6].002 Operator: Paulo Serra Run Date: 2013-04-02 21:25 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments 47 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Figure 3.8 - Thermal analysis for [C3C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. [C4C1im][PF6] was liquid at room temperature and solid phase was not reached in the measured temperature range. The glass transition was detected at 195 K (-78 ºC). Figure 3.9 - Thermal analysis for [C4C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. 46.31min 37.91°C 57.76J/g -40.93°C 139.97min 38.11°C 58.16J/g -41.21°C 233.44min 38.12°C 58.23J/g -41.15°C -100 -50 0 50 100 150 Temperature (°C) -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI249_C3C1im[PF6] Size: 11.6300 mg Method: Standard_IL_-90_150_5 Comment: AI249_C3C1im[PF6] DSC File: I:...\IonicLiquids\AI249_C3C1im[PF6].002 Operator: KRu Run Date: 2013-04-03 02:08 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments -77.33°C -78.68°C -79.35°C -100 -50 0 50 100 150 Temperature (°C) -0.6 -0.4 -0.2 0.0 0.2 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI251_C4C1im[PF6] Size: 26.5500 mg Method: Standard_IL_-90_65_5 Comment: AI251_C4C1im[PF6] DSC File: I:...\IonicLiquids\AI251_C4C1im[PF6].001 Operator: PS Run Date: 2013-03-28 20:43 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 48 [C5C1im][PF6] was liquid at room temperature and the solid phase was not reached in the measured temperature range. Glass transition was detected at 199 K (-75 ºC). Figure 3.10 - Thermal analysis for [C5C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. [C6C1im][PF6] was liquid at room temperature and solid phase was not reached in the measured temperature range. Glass transition was detected at 201 K (-72 ºC). -74.94°C -74.88°C -75.18°C -100 -50 0 50 100 150 Temperature (°C) -0.6 -0.4 -0.2 0.0 0.2 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI252_C5C1im[PF6] Size: 24.2500 mg Method: Standard_IL_-90_65_5 Comment: AI252_C5C1im[PF6] DSC File: I:...\IonicLiquids\AI252_C5C1im[PF6].001 Operator: KRu Run Date: 2013-03-29 01:29 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments 49 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Figure 3.11 - Thermal analysis for [C6C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. [C7C1im][PF6] was liquid at room temperature and solid phase was not reached in the measured temperature range. Glass transition was detected at 203 K (-70 ºC). Figure 3.12 - Thermal analysis for [C7C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. -72.20°C -71.58°C -71.10°C -100 -50 0 50 100 150 Temperature (°C) -0.6 -0.4 -0.2 0.0 0.2 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI253_C6C1im[PF6] Size: 22.0900 mg Method: Standard_IL_-90_65_5 Comment: AI253_C6C1im[PF6] DSC File: I:...\IonicLiquids\AI253_C6C1im[PF6].001 Operator: PS Run Date: 2013-03-29 06:15 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments -34.52°C 11.74°C 69.67°C -69.81°C -70.71°C -24.33°C 10.98°C 72.04°C -70.71°C -21.09°C 9.58°C 69.32°C -100 -50 0 50 100 150 Temperature (°C) -0.20 -0.15 -0.10 -0.05 0.00 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI407_C7C1im[PF6] Size: 5.9300 mg Method: Standard_IL_-90_150_5 DSC File: I:...\IonicLiquids\AI407_C7C1im[PF6].001 Operator: PS Run Date: 2013-05-09 14:42 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 50 [C8C1im][PF6] was liquid at room temperature and solid phase was not reached along the studied temperature range. Glass transition was detected at 203 K (-70 ºC). The thermogram of this ionic liquid is not presented due to an occurred error in the data storing. The following is the [C9C1im][PF6]. It was liquid at room temperature before measurements. Was observed the cold crystallization at 252 K (-21 ºC) and the liquid phase was reached at 292 K (19 ºC). Glass transition was detected at 205 K (-67 ºC). Figure 3.13 - Thermal analysis for [C9C1mim][PF6] performed from 183 to 423 K, at a 5 K∙min-1 rate. The following IL is the [C10C1im][PF6]. It was solid at room temperature and liquid phase was reached in the measured temperature range, at 307 K (34 ºC). Cold crystallization was not reached in the first measurement appearing in the two measurements after. In the first measurement a solid-solid transition at 241 K (-32 ºC)is detected. Glass transition was detected at 208 K (-65 ºC). 35.80min -18.81°C 39.77J/g -67.18°C 42.79min 19.75°C 48.10J/g 131.29min -23.63°C 38.45J/g 138.64min 19.75°C 47.99J/g 226.83min -22.75°C 38.62J/g -67.48°C -66.77°C 234.35min 19.79°C 48.08J/g -100 -50 0 50 100 150 Temperature (°C) -2.0 -1.5 -1.0 -0.5 0.0 0.5 Heat Flow (W/g) 050 100 150 200 250 300 Time (min) Sample: AI408_C9C1im[PF6] Size: 10.1400 mg Method: Standard_IL_-90_150_5 DSC File: I:...\IonicLiquids\AI408_C9C1im[PF6].001 Operator: VlkO Run Date: 2013-05-09 19:31 Instrument: DSC Q1000 V8.2 Build 268 Exo Up Universal V4.0C TA Instruments 57 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives However, for the ionic liquids measured, the studied temperature range is short and only a linear fit was considered: ,( / ) l pm C T K a bΤ (3.4) where a is in J∙K-1∙mol-1 and b is in J∙K-2∙mol-1. The fit parameters (a and b) were calculated from the step method. The mean deviation, σ, was calculated for the heat capacity results obtained from the step method. Table 3.4 - Linear fit parameters for all the studied compounds, determined from step method values. Compound Phase a (J∙K-1∙mol-1) b (J∙K-2∙mol-1) σ1 [Bnmim][BF4] Solid 6.2 0.9665 1.01 [Bnmim][BF4] Liquid 230.5 0.5436 0.05 [Bnmim][PF6] Solid 87.1 0.9058 0.76 [Bnmim][C2F4HSO3] Solid -204.2 1.7502 1.69 [Bnmim][C2F4HSO3] Liquid 322.8 0.6317 0.20 [Bnmim][NTf2] Liquid 448.5 0.5402 1.35 [C2C1im][PF6] Solid -47.3 1.2683 1.04 [C2C1im][PF6] Liquid 199.7 0.4759 0.05 [C3C1im][PF6] Solid -25.3 1.3331 0.17 [C3C1im][PF6] Liquid 230.2 0.4883 0.26 [C4C1im][PF6] Liquid 260.5 0.5070 0.64 [C5C1im][PF6] Liquid 257.4 0.6008 0.46 [C6C1im][PF6] Liquid 283.2 0.6217 0.57 [C7C1im][PF6] Liquid 305.6 0.6549 1.00 [C8C1im][PF6] Liquid 334.2 0.6773 0.67 [C9C1im][PF6] Liquid 358.6 0.7052 0.93 [C10C1im][PF6] Solid 216.8 0.8483 2.99 [C10C1im][PF6] Liquid 385.9 0.7308 0.49 [C12C1im][PF6] Solid 180.2 0.9606 1.10 [C12C1im][PF6] Liquid 392.9 0.8587 0.27 1   1/2 2 exp ,m ,m 1() nlf pp i iC C n m          , where n is the number of fitted data points and m is the number of adjustable parameters. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 58 3.2.1. Heat capacity of 1-benzyl-3-methylimidazolium hexafluorophosphate Figure 3.19 presents the deviations from the fitting data obtained, step measurements and continuous measurements for [Bnmim][PF6]. The measurements were performed from 253 to 355 K at 0.3 K∙min-1. The results presented are the average from the measurements performed. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3 and 3.4). The fitting parameters were determined from step method measurements. 250 275 300 325 350 -2 -1 0 1 2 100·(Cexp p-Clf p)/Clf p) T/K Figure 3.19 - Relative deviation of the experimental heat capacities Cpexp for [Bnmim][PF6] from the smoothed values Cplf. The black line represents the continuous method values and the squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). The experimental values of the molar heat capacity for [Bnmim][PF6] in the range 278 to 355 K and deviations are given in Table 3.5. Each value corresponds to the average of two independent measurements, using step and continuous methods. The deviation from the quadratic fit is presented for both methods. The parameters a, b and c were obtained from the fitting of the step method values (Csp,m (T) = 153.7 -0.882∙T+5.570∙10-5∙T2 from quadratic fit and Csp,m (T)= 87.1+0.9058∙T for linear fit). 59 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Additionally, the effect of water on the heat capacities measurements was evaluated for the studied ionic liquid. Table 3.5 lists the molar heat capacities for dried and the same sample before drying. Table 3.5 - Experimental molar heat capacities (J∙K-1∙mol-1) for solid [Bnmim][PF6] (dried and not dried sample) and the data derived from quadratic fitting for both step and continuous method. Dried Sample Not dried sample Step method Continuous method Step method Continuous method T/K Cp,m Quadratic fit % dev. Cp,m % dev. Cp,m % dev. Cp,m % dev. 258.09 321.08 320.75 0.10 321.68 0.21 317.55 -1.06 318.51 -0.76 263.20 324.93 325.42 -0.15 326.45 0.27 321.69 -1.19 323.60 -0.59 268.30 329.76 330.10 -0.10 331.03 0.27 326.46 -1.19 328.37 -0.60 273.41 335.03 334.77 0.08 336.25 0.46 331.55 -0.92 334.42 -0.11 278.51 339.71 339.45 0.08 339.94 0.20 336.33 -0.88 338.55 -0.23 283.62 344.65 344.13 0.15 344.18 0.10 341.10 -0.83 343.33 -0.16 288.72 348.73 348.81 -0.02 348.21 -0.06 345.69 -0.88 348.42 -0.04 293.83 353.16 353.49 -0.09 352.52 -0.12 349.69 -0.94 353.51 0.14 298.93 358.03 358.16 -0.04 356.30 -0.35 354.15 -0.94 357.33 -0.10 304.04 362.79 362.84 -0.02 360.70 -0.39 358.92 -0.90 360.83 -0.35 309.14 367.39 367.52 -0.04 364.92 -0.48 363.37 -0.86 363.37 -0.87 314.25 372.53 372.20 0.09 369.82 -0.38 368.46 -0.78 367.19 -1.10 319.35 377.53 376.87 0.17 374.87 -0.24 373.56 -0.59 371.65 -1.07 324.46 381.58 381.55 0.01 379.03 -0.35 377.37 -0.76 375.78 -1.15 329.56 385.86 386.23 -0.10 383.75 -0.30 381.51 -0.94 379.92 -1.33 334.67 390.32 390.90 -0.15 389.15 -0.09 385.65 -0.97 385.01 -1.12 339.77 395.05 395.58 -0.14 394.23 0.05 390.10 -0.96 389.78 -1.07 344.88 399.94 400.26 -0.08 399.77 0.29 394.56 -1.07 394.24 -1.12 349.98 405.06 404.94 0.03 404.77 0.40 399.01 -0.99 398.69 -1.12 355.09 410.45 409.61 0.20 410.81 0.76 403.46 -1.03 403.78 -0.90 Figure 3.20 depicts the relative deviation of the experimental heat capacities and the values obtained from the fitting, for a sample not dried. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 60 250 275 300 325 350 -2 -1 0 100·(Cexp p-Clf p)/Clf p) T/K Not dried step method Not dried continuous method Figure 3.20 - Relative deviation of the experimental heat capacities Cpexp for [Bnmim][PF6] not dried, from the smoothed values Cplf. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). When compared with the dried sample values, it was found that the heat capacity values are lower in the presence of water, as shown in Figure 3.21 (the water content was lower than 1% before drying). In blue and in green are represented the values for the same sample but not dried (in blue the continuous method and in green the step method). “0 line” stands for the values obtained by quadratic fitting. 250 275 300 325 350 -2 -1 0 1 100·(Cexp p-Clf p)/Clf p) T/K Step Continuous Not dried step Not dried continuous Figure 3.21 - Relative deviation of the experimental heat capacities, Cpexp, for [Bnmim][PF6] from the fit values, Cplf, for both dried and not dried samples. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). 61 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.2.2. Heat capacity of 1-benzyl-3-methylimidazolium tetrafluoroborate Figure 3.22 presents the fitting data obtained from step and continuous measurements for [Bnmim][BF4]. The measurements were performed from 253 to 355 K at 0.3 K∙min-1. The results presented are the average from the measurements performed. A phase transition was detected at 336 K represented by the dashed vertical line. “0” stands for the values obtained by quadratic fit (parameters are presented in table 3.3 and 3.4). The fitting parameters were determined from step method measurements. 250 275 300 325 350 -2 -1 0 1 2 100·(Cexp p-Clf p)/Clf p) T/K Figure 3.22 - Relative deviation of the experimental heat capacities Cpexp for [Bnmim][BF4] from the smoothed values Cplf. The black line represents the continuous method values and the squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). The experimental values of the molar heat capacity for [Bnmim][BF4] in the range 278 to 355 K and deviations are given in Table 3.6. Each value corresponds to the average of two independent measurements, using step and continuous methods. The deviation from the quadratic fit is presented for both methods. The parameters a, b and c were obtained from the fitting of the step method values (Csp,m (T) = 545.4-6.296∙T+1.217∙10-2∙T2 from quadratic fit and Clp,m (T)= 230.5+0.5436∙T and Csp,m (T)= 6.2+0.9665∙T for linear fit). The heat capacity values measured in the temperature range of 258 to 278 K and 334 to 339 K are not presented due to phase transitions occurring in those temperature ranges. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 62 Table 3.6 - Experimental molar heat capacities (J∙K-1∙mol-1) for solid and liquid phases of [Bnmim][BF4] and results derived from fitted equation for both step and continuous method. Step method Continuous method T/K Cp,m Quadratic fit % dev. Cp,m % dev. 258.09 - - - - - 263.20 - - - - - 268.30 - - - - - 273.41 - - - - - 278.51 288.03 287.24 0.28 295.05 1.96 283.62 292.27 291.69 0.20 298.34 2.18 288.72 296.71 296.32 0.13 301.33 1.99 293.83 301.12 301.11 0.00 302.58 1.17 298.93 305.96 306.08 -0.04 303.72 0.06 304.04 310.73 311.21 -0.15 307.82 -0.56 309.14 316.25 316.53 -0.09 315.20 -0.15 314.25 322.04 322.01 0.01 323.99 0.58 319.35 327.54 329.66 -0.64 331.66 -0.76 324.46 334.79 333.49 0.39 337.09 0.08 329.56 - - - - - 334.67 - - - - - 339.77 - - - - - 344.88 417.85 417.93 -0.02 417.44 -0.12 349.98 420.89 420.71 0.04 419.96 -0.22 355.09 423.40 423.48 -0.02 423.30 -0.04 63 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.2.3. Heat capacity of 1-benzyl-3-methylimidazolium 1,1,2,2tetrafluoroethanesulfonate Figure 3.23 presents the fitting data obtained from step and continuous measurements for [Bnmim][C2F4HSO3]. The measurements were performed from 253 to 355 K at 0.3 K∙min-1. The results presented are the average from the measurements performed. A phase transition was detected at 314 K represented by the dashed vertical line. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3 and 3.4). The fitting parameters were determined from step method measurements. Continuous method data was removed due to an error occurring along the measurements. 250 275 300 325 350 -2 -1 0 1 2 100·(Cexp p-Clf p)/Clf p) T/K Figure 3.23 - Relative deviation of the experimental molar heat capacities Cpexp for [Bnmim][C2F4HSO3] for step method. The squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 64 The experimental values of the molar heat capacity for [Bnmim] [C2F4HSO3] in the range 278 to 355 K and deviations are given in Table 3.7. Each value corresponds to the average of two independent measurements, using step and continuous methods. The deviation from the quadratic fit is presented for both methods. The parameters a, b and c were obtained from the fitting of the step method values (Csp,m (T) = 1068.4 -5.319∙T+1.266∙10-2∙T2 and Clp,m (T)= 507.0 -0.470∙T +1.590∙10-3∙T2 for quadratic fit and Csp,m (T)= -204.2 +1.7502∙T and Clp,m (T)= 322.8 +0.6317∙T for linear fit). The heat capacity values measured in the temperature range of 258 to 268 K and 314 to 319 K are not presented due to phase transitions occurring in those temperature ranges. Table 3.7 - Experimental molar heat capacities (J∙K-1∙mol-1) for solid and liquid phases of [Bnmim][C2F4HSO3] and results derived from fitted equation for both step and continuous method. Step method Continuous method T/K Cp,m Quadratic fit % dev. Cp,m % dev. 258.09 - - - - - 263.20 - - - - - 268.30 - - - - - 273.41 380.59 379.77 0.22 392.30 3.30 278.51 388.59 389.46 -0.22 398.39 2.29 283.62 398.36 399.62 -0.32 405.78 1.54 288.72 407.11 410.22 -0.76 415.37 1.26 293.83 422.95 421.30 0.39 423.94 0.63 298.93 436.35 432.81 0.82 434.12 0.30 304.04 441.43 444.80 -0.76 447.89 0.70 309.14 456.94 457.22 -0.06 - - 314.25 - - - - - 319.35 - - - - - 324.46 521.83 521.50 0.06 526.72 1.00 329.56 524.50 524.62 -0.02 529.72 0.97 334.67 527.44 527.75 -0.06 526.72 -0.20 339.77 530.80 530.87 -0.01 529.72 -0.21 344.88 533.96 534.00 -0.01 532.99 -0.19 349.98 537.23 537.12 0.02 537.01 -0.02 355.09 540.35 540.26 0.02 540.70 0.08 65 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.2.4. Heat capacity of 1-benzyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide. In Figure 3.24 is presented the fitting data obtained from step and continuous measurements for [Bnmim][NTf2]. The measurements were performed from 253 to 355 K at 0.3 K∙min-1. The results presented are the average from the measurements performed. “0” stands for the values obtained by quadratic fit (parameters are presented in table 3.3 and 3.4). The fitting parameters were determined from step method measurements. 250 275 300 325 350 -2 -1 0 1 2 100·(Cexp p-Clf p)/Clf p) T/K Figure 3.24 - Relative deviation of the experimental molar heat capacities Cpexp for [Bnmim][NTf2] for step method. The squares are step method results. “0 line” stands for the values obtained by quadratic fit (parameters are presented in table 3.3). FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 66 The experimental values of the molar heat capacity for [Bnmim][NTf2] in the range 278 to 355 K and deviations are given in Table 3.8. Each value corresponds to the average of two independent measurements, using step and continuous methods. The deviation from the quadratic fit is presented for both methods. The parameters a, b and c were obtained from the fitting of the step method values (Clp,m (T) = 599.8 -0.148∙T+1.150∙10-3∙T2 from quadratic fit and Clp,m (T)= 448.5 +0.5402∙T for linear fit). Table 3.8 - Experimental molar heat capacities (J∙K-1∙mol-1) for liquid [Bnmim][NTf2] and results derived from fitted equation for both step and continuous method. Step method Continuous method T/K Cp,m Quadratic fit % dev. Cp,m % dev. 258.09 587.80 585.33 0.42 588.65 0.19 263.20 588.75 588.16 0.10 590.85 -0.13 268.30 591.75 590.99 0.13 593.41 -0.17 273.41 594.37 593.82 0.09 599.39 -0.15 278.51 596.27 596.65 -0.06 597.51 -0.21 283.62 599.16 599.49 -0.06 599.69 -0.28 288.72 601.32 602.32 -0.17 601.86 -0.38 293.83 603.68 605.15 -0.24 604.13 -0.46 298.93 606.86 607.98 -0.18 606.39 -0.40 304.04 609.81 610.81 -0.16 609.02 -0.38 309.14 612.46 613.64 -0.19 611.43 -0.40 314.25 615.97 616.47 -0.08 615.08 -0.29 319.35 619.54 619.30 0.04 618.40 -0.17 324.46 621.92 622.13 -0.03 620.92 -0.24 329.56 624.68 624.97 -0.05 623.45 -0.25 334.67 627.43 627.80 -0.06 626.55 -0.26 339.77 630.49 630.63 -0.02 629.83 -0.22 344.88 634.11 63346 0.10 633.12 -0.09 349.98 637.55 636.29 0.20 636.32 0.00 355.09 640.60 639.12 0.23 640.07 0.04 73 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives Figure 3.31, presents a comparison between specific heat capacities of the [CnC1im][NTf2] series reported by Rocha et al [41], and the results obtained in this work for the [CnC1im][PF6] series. The PF6 series presents higher specific heat capacity than the NTf2 Series. 3 5 7 9 11 1.3 1.4 1.5 1.6 1.7 [CnC1im][PF6] [CnC1im][NTf2] co p (J·K-1·g-1) n [C] Figure 3.31 - Specific heat capacities, at T = 298.15 K, as function of the number of carbon atoms in the alkyl side chain of the cation, n(C), for [CnC1im][PF6] (with n = 2 – 8, 10, 12) and [CnC1im][NTf2] (with n = 2 – 8, 10, 12). In Figure 3.32, the volumic heat capacity dependence on the number of carbons of the alkyl side chain of the cation for the [CnC1im][PF6] and [CnC1im][NTf2] ionic liquid series is represented. The volumic heat capacities, Cp /V, for the [CnC1im][PF6] were calculated taking into account the specific heat capacities and the density data available in the literature. [40] Cp /V data for the [CnC1im][NTf2] ionic liquid series was recently reported in the literature by Rocha et al [39, 42]. The results will be evaluated based on a comparative analysis with the [CnC1im][NTf2] in order to explore the effect of the anion on the heat capacity data. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 74 A trend shift around C6 and C7 was found along the volumic heat capacity with the alkyl side chain length for PF6. For shorter alkyl chain length size, the volumic heat capacity decreases with the increase of the alkyl chain size. After C7, the volumic heat capacities reaches a stationary value of 1.94 J∙K-1∙cm-3, slightly higher than the identical behaviour observed in the [CnC1im][NTf2] IL series (1.92 J∙K-1∙cm-3)[42]. 2 4 6 8 10 12 1.90 1.92 1.94 1.96 1.98 2.00 [CnC1im][NTf2], n=3,4 and 5 [CnC1im][NTf2], n=7,8,10 and 12 [C6C1im][NTf2] [C2C1im][NTf2] [C4C1im][PF6] [CnC1im][PF6], n=5,6,7,8,9 Co p,m/V (J·K-1·cm-3) n[C] Figure 3.32Volumic heat capacities at 298 K, as function of the number of carbon atoms in the alkyl side chain of the cation 75 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.3. Comparison of experimental data with estimation methods 3.3.1. Group contribution methods for phase transitions In Table 3.11, comparison of experimental glass and melting temperatures obtained in this work with the group contribution method developed by Lazzús [34, 43] for Tg and Tm, represented in eq. 3.5 and 3.6, respectively.   178.63 g i ci j aj T K n Δt n Δt    (3.5)   288.70 m i ci j aj T K n Δt n Δt    (3.6) Literature experimental data reported by Chun et al [44] is also reported to support the obtained results. Table 3.11 - Tg and Tm experimental results obtained from [Bnmim][BF4], [Bnmim][PF6], [Bnmim][C2F4HSO3], [Bnmim][NTf2] and [CnC1im][PF6], where n = 2 - 10 and 12. And comparison with the estimated values computed from Lazzús group contribution method [34, 43] and literature data from Chun et al [44]. IL Tg (K) Tm (K) Calculated Experimental Literature[44] Calculated Experimental [Bnmim][BF4] 467.4 235.3 - 360.0 336.5 [Bnmim][PF6] 474.4 244.8 - 335.8 399.7 [Bnmim][C2F4HSO3] 397.4 229.1 - 271.6 315.4 [Bnmim][NTf2] 445.7 216.7 - - - [C2C1im][PF6] 194.0 - - 357.9 317.0 [C3C1im][PF6] 195.1 - - 354.2 311.8 [C4C1im][PF6] 196.3 196.0 196.2 350.4 - [C5C1im][PF6] 197.5 199.2 193.2 346.6 - [C6C1im][PF6] 198.6 201.7 193.2 342.9 - [C7C1im][PF6] 199.8 203.5 189.2 339.1 - [C8C1im][PF6] 200.9 203.5 202.2 335.4 - [C9C1im][PF6] 202.1 205.9 207.2 331.6 293.0 [C10C1im][PF6] 203.2 208.2 - 327.8 307.1 [C12C1im][PF6] 205.6 - - 324.1 326.5 FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 76 Calculated values for [Bnmim] series are highly influenced by the values for benzyl (274.990 K) and for SO2- (-759.500 K) which the author, Lazzús [34, 43] confirmed. Those values deeply influence the results computed (notice that for if a IL with SO2-, it is impossible to possess a positive melting temperature or glass transition) and do not allow an accordance with the values for the experimental data obtained. Obtained results are, in general, in agreement with the estimated values with exception of [Bnmim] Tg values where the value for benzyl group seems rather high and influences the Tg values computed for this set of compounds. It was also observed that the Tg and the heat capacities, in compounds such as ILs, increase with the number of atoms in the alkyl chain, in accordance with Paulechka [45]. In Table 3.12, another GCM was used, from Gharagheizi et al [46, 47], but the Tg (computed through eq. 3.7 for Tg and eq. 3.8 for Tm) values are not in accordance with the experimental data. Also, there is a big lack of groups and when a group is presented, evaluation is very difficult due rather complex definitions presented in the paper by Gharagheizi et al [46, 47], which can easily lead to error.   158.87 g i ci j aj T K n Δt n Δt    (3.7)   264.29 m i ci j aj T K n Δt n Δt    (3.8) Table 3.12 - Tg and Tm experimentally obtained for [Bnmim][BF4], [Bnmim][PF6], [Bnmim][C2F4HSO3], [Bnmim][NTf2] and the estimated values from Gharagheizi et al [46, 47] GCM. IL Tg (K) Tm (K) Computed Experimental Calculated Experimental [Bnmim][BF4] - 235.3 - 336.5 [Bnmim][PF6] - 244.8 - 399.7 [Bnmim][C2F4HSO3] 289.8 229.1 311.9 315.4 [Bnmim][NTf2] 335.1 216.7 - - 77 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 3.3.2. Group contribution method for heat capacities In Table 3.13 are presented the heat capacities for the [CnC1im] series, at 298.15 K, and show the influence of the alkyl chain growth in the heat capacity. The experimental data was compared with the group contribution method (GCM) by Gardas et al [27] represented in eq. 3.9. 2 ( / ) [ ( /100) ( /1000) ] p C T K R A B T D T   (3.9) And literature data from Paulechka (for C4C1) [45]. Table 3.13 - Experimental molar heat capacity data, at 298 K, obtained for [CnC1im][PF6], where n = 2 - 10 and 12, and comparison with Gardas et al [27] group contribution method and literature data from Paulechka [45]. [CnC1im][PF6] Cp exp (J∙K-1∙mol-1) CpGCM (J∙K-1∙mol-1) Cpliterature (J∙K-1∙mol-1) [C2C1im][PF6] 346.9 ± 0.2 343.6 - [C3C1im][PF6] 377.5 ± 0.4 375.6 - [C4C1im][PF6] 411.6 ± 0.9 407.6 408.1 [C5C1im][PF6] 435.6 ± 0.1 439.6 - [C6C1im][PF6] 467.4 ± 0.1 471.6 - [C7C1im][PF6] 498.9 ± 0.1 503.6 - [C8C1im][PF6] 534.8 ± 0.0 535.6 - [C9C1im][PF6] 567.7 ± 0.1 567.6 - [C10C1im][PF6] 600.8 ± 12.8 599.6 - [C12C1im][PF6] 659.5 ± 2.0 663.6 - In figure Figure 3.33 presents the experimental results and the comparison with the GCM computed values presented by Gardas et al [27] and a literature value. C2, C4, C10 and C12 are solids at 298 K. For those ILs, the liquid heat capacities, at 298 K, were derive by linear extrapolation from the liquid region of the experimental data. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 78 2 4 6 8 10 12 200 400 600 800 Co p,m / J·K-1·mol-1 n[C] Experimental data Group contribution method values Literature data Figure 3.33 - Comparison between the data obtained for [CnC1im][PF6], where n = 2 - 10 and 12, and the data from group contribution method and literature data found for the same compounds. In Figure 3.34 is presented the relative deviation of the experimental values from the GCM values. The graphic show good correlation between experimental values and GCM computed values. 2 4 6 8 10 12 -2 -1 0 1 2 Experimental values Literature values 100.(Cexp p,m-Clf p,m)/Clf p,m) n[C] Figure 3.34 – Relative deviation from GCM values (0 line), of the experimental and literature values for [CnC1im][PF6], where n = 2 - 10 and 12. “0 line” stands for the values obtained by linear fit. 79 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 80 4. Conclusions 81 FCUP Thermal behavior and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives This work was focused on the thermal study, condensed phases heat capacities measurements and phase behaviour studies of ionic liquids. Two families were studied. One based on the 1-benzyl-3-methylimidazolium cation conjugated with 4 anions (tetrafluoroborate; hexafluorophosphate; 1,1,2,2-tetrafluoroethanesulfonate; bis(trifluoromethylsulfonyl)imide), and one based in the 1-alkyl-3-methylimidazolium series: - methyl; -ethyl; -propyl; -butyl; -pentyl; -hexyl; –heptyl; -octyl; -nonyl; -decyl and -dodecyl-3methylimidazolium with the anion hexafluorophosphate. A comparative analysis for the anion effect in the 1-benzyl-3-methylimidazolium ionic liquids, and the benzyl group effect was performed, as well as the alkyl side chain length effect on the properties of 1-alkyl-3methylimidazolium series. The thermal phase behaviour study was carried out in the temperature range 183 to 423 K, and compared with available literature values as well as with group contribution estimation methods. The results from TA Instruments Q1000 DSC were compiled and shown in Table 3.1 and the ratio Tg/Tm was calculated. The Tg/Tm ratio of the [Bnmim][PF6] the was found to be outside the typical [2/3 to 3/4 ] range. Experimental glass transition temperatures Tg obtained in this work are well predicted by the method of Lazzús [34, 43] where it can be found that for 1-alkyl3-methylimidazolium, Tg results are typically in the region between 183 K and 203 K in accordance with Holbrey et al [48]. For melting temperatures Tm the deviations from the prediction by GCM is significantly due to the particular effect of the structuration and nanostructuration of the ILs in the relative stability between crystal and the liquid phases. From the thermal behaviour results of the PF6 series, some indication was found concerning the relation between the solid-solid transition temperature, Ts-s, and the appearance of the cool crystallization temperature Tcc. The thermal analysis profile of studied ILs series could be divided in two distinct regions/behaviours. The region is characterized by a small decrease in the Tm and Ts-s, along the alkyl side chain, and starting from [C7C1im][PF6] an increase of the Tm and Ts-s is observed with the alkyl side chain length in agreement with the trend shift recently reported in the literature for several physicochemical properties along the ILs series [39, 42]. Condensed state heat capacities data are reported, in the temperature range 258 to 355 K, and compared with available literature values as well as with group contribution estimation methods. FCUP Thermal behaviour and heat capacity of ionic liquids: benzilimidazolium and alkylimidazolium derivatives 82 Since the data in the literature for the studied ILs is scarce, estimation methods were used to support experimental data. The experimental heat capacities are quite well predicted by the Gardas et al [27] group contribution method.