Home

LEITAO_THESIS - Repository TU Delft

1. panel 134 This is an extensively versatile and powerful piece of software which allows for the visualization and extraction of any two sets of data measured by any of the modules connected to the control PC during a measurement As such its adequate use enables not only the extraction of the data necessary for the calculation of specific heat but also the detailed monitoring of all the events which might have happened with the different modules during a measurement A 3 3 Calculation software These consist of two Matlab routines which do not need to be run in the same computer onto which the three Keithley modules and the Lakeshore 331 Temperature Controller are connected via a GPIB bus but merely in one where Matlab is installed The first of these routines entitled grabtau m uses two data sets that are extracted from your measurement tdms file the Keithley 2000 voltage reading vs time and the Keithley 2400 voltage reading vs time and it calculates the specific heat ratio between the sample and reference in order of time The second routine entitled capacity temperature matcher m resorts to an additional data set the Lakeshore 331 Temperature Controller sample space temperature reading vs time and it matches the specific heat ratio calculated by the grabtau m routine with temperature Both these routines were written by Peter van Dommelen as part of his Bachelor end project and they are shown below w
2. 47 5 1 Introduction 9 LLL 47 5 2 Material overview nennen 48 5 2 1 The Mn Fe Sisystem 48 32 2 IheMnis Pe P System odores SU 5 3 Sample preparation and characterization L 5 5 4 Results and discussion s C O 5 5 4 1 Structural results SW 5 4 2 Magnetic results DA 5 5 The Mn Fe 3 Si P magnetostructural phase diagram 1 57 5 5 1 Overall diagram description 539 5 6 Mn Fe 3 Si P B compounds 62 5 6 1 Sample preparation and characterization 3 5 6 2 Results and discussion 68 5 6 3 Magneticresults 5 57 CoD USOS unm oeste tc nen 67 References 68 SYSICHIS c or p fox orta E a ado Ee e tati rerum rep 6 1 Introduction LL LL LLL 7l 6 2 Material overview cos os dennen dte nde tented 7l 6 2 1 The Fe Co Si P system TL 6 2 2 The Mn Co Si P system TD vi 6 3 Results on the Fe Co 3 Si P system 6 3 1 S rich samp GL noren EEn 6 3 2 P nch samples srt ee ee 6 3 3 Sulfur substitution in Fe Co 3 P Si 6 4 Results on the Mn Co 3 Si P system 6 4 1 Inverse magnetocaloric effect 6 5 Conclusions References Chapter 7 Magnetic and Magnetocaloric Exploration of Fe rich Mn Fe P Ge 7 1 Introduction 7 3 Sample preparation and characterization 7 4 Result and discussion TAI Mn Be BGO system EE 7 4 2 Mn Fe 9s P Ge system 7 5 Permanent magnet potential 7 6 Conclusions References Chapter 8 In field microcalorimety measurements on polycrystalline Fe P 8
3. oo x 9 A 9 4 lt x lt 9 9 lt x lt 00 lt 100 gt Hard Hard Medium hard lt 110 gt Medium hard Easy Easy lt 111 gt Easy Medium hard Hard 2 2 3 Estimation of anisotropy constants The determination of the values of the anisotropy constant values can be performed in various ways 16 A straightforward method is by measuring the anisotropy field Han This is achieved by measuring magnetization curves with a magnetic field applied parallel and perpendicular to the easy magnetization direction The intersection of these two curves then gives us the anisotropy field 3 An example of such a measurement is given in Figure 2 4 Sm Co cFeo 4 47 TERT EN a ei SmCos single crystal Il c axis Intrinsic induction B T SmCos 1 c 0 10 20 30 External field oH T Figure 2 4 Several easy hard magnetization curves for various compounds 7 Intrinsic induction being 44M Supposing that a material s spontaneous magnetization Ms has an angle 6 with a field applied perpendicular to the symmetry axis the magnetic field will exert a torque of 44 HM cos0 which will tend to increase with The value of this torque can be obtained by differentiating the expression for the anisotropy energy df y do 2K sin cos 0 L AK sin 6 cos 0 2 18 Taking the two torques into consideration we arrive at the following expression 2K sin 4K sin 8 M mH 2 19 S 17 Knowing that the va
4. 60 40 Magnetization Am kg 20 0 50 100 150 200 250 300 2350 400 Temperature K 140 120 nn o a E ERN 100 E E 80 vi O 60 N z 40 4 Tesla 20 Mn Fe P Ge NS i y Mn Fe 3PosGe s 0 50 100 150 200 250 300 350 400 Temperature K Figure 7 5 a Comparison between the Ferromagnetic and the anomalous magnetic behavior observed in low Ge content samples in the Mn Fe P Ge system b Comparison between two samples with diferent Fe contents demonstrating the decrease in transition sharpenss promoted by the increase of Fe The loss of the first order behavior of the magneto elastic transition also implies a disappearance of the discontinuity previously observed in the Mn rich magnetic and structural properties This can be best observed by monitoring the 003 and 002 peaks in the X ray diffraction patterns of the Fe rich Mn Fe P Ge samples 93 While in the Mn rich side of the diagram we observe a discontinuity in the temperature evolution of these two peaks around the transition temperature exemplified in Figure 7 3 we now observe a very smooth and gradual variation with temperature in both of them as seen in Figure 7 6 460 450 0 88 440 0 72 430 0 56 420 0 40 O 5 410 0 24 400 0 2 390 5 380 370 360 LA 350 51 52 53 54 55 Position 20 Copper Figure 7 6 Contour plot of the angle change in the 300 211 and 002 peaks versus
5. 1 linlogmax linlogmax polyval P pifstarttime index 1 else linlogmax linlogmax polyval P data time downwards index 1 end high high mean data abs downwards index 1 5 downwards index 1 J error error mean polyval P timedown logdown 2 Serror used in showing worst fit end Calculate initial voltage and decay time linmax exp linlogmax inverse of the logarithm corrA linmax 1 2 end corrB linmax 2 2 end if fitorder 1 slopeA resultdown 1 2 end 1 slopeB resultdown 2 2 end 1 else slopeA 2 resultdown 1 2 end 1 data time downwards 1 2 end 1 nSkip resultdown 1 2 end 2 138 slopeB 2 resultdown 2 2 end 1 data time downwards 2 2 end 1 nSkip resultdown 2 2 end 2 end decayA 1 slopeA decayB 1 slopeB highA high 1 2 end highB high 2 2 end temp corr A corrA seebeck corrB seebeck temp corr B Calculate average results heat_resistance_correction_A corrA highA heat resistance correction B corrB highB tance A heat resis temp corr A approx power 1 heat_resistance_B temp_corr_B approx_power valA decayA heat_resistance_A valB decayB heat resistance B moving valB 0 5 valB l1 end 1 valB 2 end ratio valA 2 end 1 moving valB average mean ratio std devi
6. m N GQ A a e 500 1000 1500 2000 2500 Time s Figure 4 6 Specific heat data as acquired by the Keithley 2000 The alternating hight of the relaxation cycles is due to the alternating readout between the sample and the reference chip The calculated relaxation is shown in Figure 4 7 For this calculation certain particularities of the measurement system need to be taken into consideration such as an inbuilt instantaneous temperature jump that may occur in the sensor temperature readout of the microcalorimetry chips To correct for this problem the fit is only 41 performed between two movable start and finish points removing the initial points of each relaxation curve from our calculation Measurement Fit fit range Logarithm of the excitation voltage T2 74 76 78 80 82 Time s Figure 4 7 Linear fit of the logarithm of the thermopile output while relaxing From the calculated relaxation time and the initial voltage supplied by the Keithley 2400 the ratio of the total heat capacity between the two chips can be calculated Knowing the masses and molar weight of both samples together with the specific heat of the reference this value can further worked until the heat capacity of the sample material is obtained as displayed in Figure 4 8 42 0 3 Tesla 160 from ref 23 200 220 230 240 250 Temperature K Figure 4 8 Isolated Fe P specific heat measurement over time for an a
7. n o sei Porqu n o sei ainda H sempre qualquer coisa que est pra acontecer Qualquer coisa que eu devia perceber Porqu n o sei Porqu n o sei Porqu n o sei ainda Ensinas me fazer tantas perguntas Na volta das respostas que eu trazia Quantas promessas eu faria Se as cumprisse todas juntas N o largues esta m o no torvelinho Pois falta sempre pouco para chegar Eu n o meti o barco ao mar Pra ficar pelo caminho C dentro inqueitacdo inquieta o E s inquieta o inquieta o Porqu n o sei Porqu n o sei Porqu n o sei ainda H sempre qualquer coisa que est pra acontecer Qualquer coisa que eu devia perceber Porqu n o sei Porqu n o sei Porqu n o sei ainda C dentro inqueitag o inquieta o s inquieta o inquieta o Porqu n o sei Mas sei que n o sei ainda H sempre qualquer coisa que eu tenho que fazer Qualquer coisa que eu devia resolver Porqu n o sei Mas sei Que essa coisa que linda Jos M rio Branco
8. F Smith G K Bristow C V Boughton and P R W Hudson J Phys F Metal Phys 10 1980 2867 H C Xuan D H Wang C L Zhang Z D Han B X Gu and Y W Du Appl Phys Lett 92 2008 102503 S H Xie J Q Li Y H Zhuang J Magn Magn Mater 311 2007 589 N T Trung First order phase transitions and giant magnetocaloric effect Delft University of Technology 2010 PhD thesis Chapter 6 69 70 Chapter 6 Magnetocaloric exploration of the Fe Co 3 Si P and Mn Co Si P systems 6 1 Introduction The current study was motivated by the discovery of the novel P6 mmm hexagonal phase in the Mn Fe 3 Si P system Its aim was to ascertain the existence of this hexagonal phase in related systems and its potential for magnetocaloric applications The selection of cobalt as both a substitute for iron and manganese was due to both its elemental proximity and affordable price The toxicity of cobalt when compared to that of manganese or iron is not irrelevant but it is lower than that of nickel or chromium our other options for this study As the hexagonal P6 mmm structure was found to be sensitive to the experimental procedure used during the sample preparation all studied samples of the Fe Co 3 Si P and Mn Co Si P systems were produced by following the same experimental procedure as used for the Mn Fe Si P system described in Chapter 5 The used elemental cobalt in our sample preparation was 99 8 pure
9. This field alignment was further verified by X ray diffraction Measurements were conducted on the random powder and on the magnetic field aligned powder of the Mn Fe gsPo gsGeo 1s sample as shown in Figure 7 12 Random powder 3000 2000 1000 2000 P Field aligned powder Counts 1500 1000 500 35 40 45 50 55 60 Position 20 Copper Figure 7 12 X Ray diffraction measurements of Mn Fe ssPo ssG amp o s The top graph shows a regular randomly aligned powder and the bottom one a field aligned powder in which it is possible to observe an increase in the intensity of the 002 reflection indicating an alignment along the c direction These X ray diffraction measurements which show a typical hexagonal Fe P structure reveal a clear increase in the 002 peak intensity relatively to the other peaks which indicates the presence of an easy axis along the c direction once again underlying the presence of magnetic anisotropy in this sample and its possible viability as a permanent magnet 101 The only drawbacks of this surprising discovery so far are the obvious lack of coercivity or a broad hysteresis loop a property which plays a crucial role in the ability of a magnet to resist thermal demagnetization and the relatively low Tc of 430 K for the Mno Fe s 5Po9Geo sample This Tc value although much higher than that of pure Fe P 28 is still too low for practical permanent magnet applications T
10. and is displayed in Figure A 15 Figure A 15 Wire layout on the D1 and D2 cables Given the number of repeated wire connection in the D1 and D2 insert plugs the red circles in the D2 cable indicate the pins which are being used in this cable 125 The D1 and D2 cables finally connect the insert plugs and the wires emerging from the sample space to the external measurement modules Observing these into the Lakeshore 331 Temperature Controller one should have two separate wires wires B and hl and a four wire cable consisting of wires 5 6 7 and 8 into the Keithley 2400 one should have two wires wires 3 and 4 into the Keithley 2002 two wires these are duplicates of wires feeding into the Keithley 2000 and are mainly used for diagnosis purposes and all remaining wires should feed into the Keithley 2000 making up channels 1 through 4 A 3 Software overview There are several programs and routines that have to be used between the steps of data acquisition and achieving a final presentable result These are listed below 1 The microcalorimetry instrument interface software Labview program 2 The data visualization and extraction software Labview program 3 The calorimetric data calculation software Matlab routine 4 Data plotting and handling software Origin or any other versatile data plotter A 3 1 Interface software This program should be installed in a computer onto which the three Keithley modules and the Lakes
11. heat peak and sharpness is detectable with increasing field as demonstrated in Figure 8 2 While this particular magnetic behavior falls in line with the results previously obtained by Caron et al 10 shown in Figure 8 1 the transition temperature did not seem to be influenced by the application of an external magnetic field maintaining a stable and strangely high value of 231 K This temperature inconsistency may be due to instrumental issues The Lakeshore 331 Temperature Controller possesses a thermal correction algorithm consisting of reversing the polarity of the current source every other reading a procedure referred to as Reversal Mode In this way the average of the positive and negative sensor readings cancel the thermal EMF electro motive force voltage enabling for more accurate measurements Unfortunately our original temperature calibration displayed in Figure 4 3 was done without resource to the Reversal Mode function which on certain temperature ranges may give rise to substantial temperature uncertainties 108 160 120 0 Tesla 80 160 1204 0 1 Tesla X 80 2 160 3 420 O 80 160 120 80 1 176 184 192 200 208 216 224 232 2240 248 Temperature K Figure 8 2 Fe P specific heat measurements taken in the experimental setup described in Chapter 4 with magnetic fields between 0 and 0 3 Tesla The specific heat values for solid copper used to calculate the contribution of our reference
12. s5FeosSio sPo5 presents values of a 8 83 and c 10 89 A The values decrease for increasing Fe content arriving at an a 8 77 A and c 10 64 for Mn Abe 6S1o sPo s X ray diffraction fitting for samples with a higher Fe content than 1 6 was found to be unreliable as these samples presented a larger amount of secondary phases belonging to cubic and orthorhombic structures The same problem occurred for our samples with other Si P ratios than the presented 0 5 0 5 a eo M ai ee 2 b i 9 7 9 OU F f 2 D Q an ITA lem Q 2 Mn Fe atoms Si P atoms Figure 5 3 a Pmmm orthorhombic structure of the Mn Fe P for values of x betewn 1 and 2 aproximatly b P6 mmm hexagonal structure of the Mn jFe Si P for 0 2 lt x lt 2 0 and 0 2 y 0 9 The borders between the various structural phases in this system were found to be significantly influenced by the annealing temperature used in their production Having tested various annealing temperatures on a set of MnFesSiosPos samples displaying cubic hexagonal and orthorhombic phases see Table 5 1 it was found that the hexagonal phase is promoted for increasing annealing temperature while the cubic 53 phase was partially suppressed The orthorhombic phase also appears to increase with temperature but not strongly 5 4 2 Magnetic results All of the samples belonging to the Pmmm orthorhombic structure in accordance with the results from Goto ef al 21
13. were found to be antiferromagnetic The novel hexagonal phase similarly to the cubic phase was found to exhibit a Ferromagnetic Paramagnetic transition above an Fe content of 1 2 This transition like those present in cubic and tetragonal phases of this system also presents the characteristics of a second order phase transition On all border regions samples were found to exhibit multiple phases In fact rigid borders or discontinuities between different phases aren t observed What was found were relatively wide bands where the bordering phases coexist Border lines can only be estimated through the analysis of phase fractions on each sample However by taking into consideration magnetic measurements a much clearer monitoring of this border may be obtained The border between the orthorhombic and tetragonal phases was mapped through the influence of the Si content on the magnetic behavior of the sample as there is a very clear difference between the antiferromagnetic orthorhombic phase and the ferromagnetic tetragonal phase Figure 5 4 Assuming that the magnetization of the Fe rich ferromagnetic tetragonal phase decreases linearly with Mn content through extrapolation of the measured values of magnetization at 5 K we can then estimate the percentage of the different phases present in our samples At 5 K we found for our Mnos5Fe 5Sio gt Pos and Mno Fes 4510 2P s samples a magnetization of 72 78 Am kg and 61 19 Am kg respectively Fo
14. 1 This will cause the Remote function of the Lakeshore 331 Temperature Controller to turn on blocking you from being able to do anything on the physical module 152 7 Change the Setpoint Set on the Lakeshore 331 Temperature Controller software block to the final temperature you wish for your ramp and press the arrow up in the Heater Range software box turning the sample space heater s on You should now observe that you are not only performing a measurement but also that the setpoint in the 331 is changing according to the ramp you defined the blue ramp light on the Lakeshore 331 Temperature Controller should be intermittent at this point while a certain heater power is being charged to the sample space heater s so as to maintain this same rate in your sample space Figure A 31 illustrates the position of the above mentioned buttons on the Lakeshore 331 Temperature Controller and its corresponding software block a EE S nud Zane Laa Display Figure A 31 a Location of all buttons mentioned in the above instruction for the performance of a temperature ramp in the Lakeshore 331 Temperature Controller front panel b Location of the Setpoint set and Heater range boxes needed to perform a remote temperature ramp 153 It should be noted that a single heater range will not be sufficient to reach the whole range of temperatures which are achievable by the setup and depending on the nature of the measuremen
15. 2 8 we thus obtain C oM dT dH 2 Zar ml 29 and consequently H T dM T H AT T d l zs all x dH 2 10 H deriving the expression for the calculation of AT 44 Observable in Equation 2 10 is the fact that besides the isothermal magnetization measurements already required in Equation 2 5 for A77 one also needs specific heat data This means that while in practical terms the use of the adiabatic process is more convenient it is in fact mathematically and experimentally more difficult 2 2 Magnetic anisotropy Anisotropy effects in magnetism relate to the existence of energetically favored spatial dimensions 2 Considering a volume V with uniform magnetization M our focus becomes the dependence of the free energy Fn m on the orientation of M Disregarding temperature dependent effects making M m with M being the saturation magnetization the state of the system is described by the magnetization unit vector as M as 2 11 er 2 11 The Cartesian components of m m m m can be expressed in terms of the spherical angles 0 and 6 by m sinOcos 2 12a m sin Osin 2 12b m cos 2 12c oe Fn The energy density simply defined as f m uni can be represented as a surface where the distance from the origin along the direction m is given by fy m An isotropic exchange generates a surface that resembles a sphere as depicted in Figure 2 1 indicati
16. 30 KQ on channels 3 and 4 These are the resistance values of the XEN 39328 chip heaters and thermopiles respectively should the chip model used not be this one please check these values with the manufacturer On the Keithley 2002 this should read as a string of unimportant numbers in channels 1 and 2 and as an Overflow on channels 3 and 4 Figure A 24 illustrates the position of the above mentioned buttons on the Keithley 2000 MCN turn off the REM emote definition Figure A 24 Location of all the buttons mentioned above in the Keithey 2000 front pannel Check the Lakeshore 331 Temperature Controller you should read a temperature value of about 300 K and a resistance value of around 12 8 Q 142 5 Should any of these values fail you have an error If this is a consistent error on either channel 1 and 3 or channel 2 and 4 then it is most likely chip related the left chip for 1 and 3 and the right chip for 2 and 4 If this is the case confirm that you have positioned the chips correctly the small elongated patch on the chip should match with the one on the socket If all seems right re examine the integrity of the chip SiN membrane and try to switch the chip positions and see if the error follows one of the chips If this happens go back to point 1 and repeat this procedure with one chip If the problem is not localized there is the possibility that both chips are malfunctioning but you are most likely facing a wiring proble
17. Detail on Image 1 produced by the routine grabtau m Table A 10 Legend for Figure A 37 1 Relaxation curves the chips thermopile voltage response as measured by the Keithley 2000 vs time 2 Results from the fits worked out from the relaxation curves for both sample and reference chip vs time 3 Worst relaxation curve fit 4 Relaxation time of all the relaxation curves for both sample and reference chip vs time 5 Heat capacity for both sample and reference chip vs time 6 Heat capacity ratio between sample and reference chip vs time 10 Run the capacity temperature matcher m 11 This should finally open one image window with the result from point 6 in Figure A 37 vs temperature 162 12 To further proceed you will need to export or copy the two vectors used in the plot made by the capacity temperature matcher m into a suitable data plotting and handling program such as Origin 13 Having the two data sets isolated in such a program you may now proceed with simple calculus steps on the appropriate columns taking into consideration sample and reference mass and specific heat in order to isolate the calorimetric values of your actual sample Arriving at your final result as exemplified in Figure A 38 160 AFe P polycrystalline 0 3 Tesla 140 120 x 100 E i gt 80 drehte a ne Ni e vh A N Manuel O sn VL TOT a 176 184 192 200 208 216 22
18. K and K can then be extracted from this graph in the same way as in the Sucksmith Thomson method as demonstrated in Figure 2 5 18 HoH uoM 1 oM T Figure 2 5 Sucksmith Thompson plot for Nd sFe Bs evidencing the extraction of the values of K and K 10 e Area Method Yet another alternative is called the Area method which has the advantage of avoiding errors due to isotropic strains In a cubic structure the variation of the anisotropy energy with the direction of magnetization is commonly expressed in terms of directional cosines Assuming OA OB and OC to be the cubic edges of the structure and magnetization direction to be OP then the anisotropy energy per volume can be given by tay Ky K a aa 050 Kla 0205 2 22 where a 2 cos POA a 2 cos POB and a cos POC The anisotropy constants can be determined by calculating the energy of magnetization as HdJ along the different crystal axes This includes the measurement of the J H curve from complete demagnetization up to magnetic saturation The area between this curve and the J axis may then be determined which adds up to the energy for each crystallographic direction 19 Taking as an example the 100 direction from Equation 2 22 we obtain a 1 anda a 0 being the end result that E K Taking the 110 direction we obtain a a and 0 From these ES 4 2 values Equation 2 22 then gives us E Ko 5K Simil
19. Multimeter used to measure the response of the Xensor microcalorimetry chip thermopiles to the square wave supplied by the Keithley 2400 the de facto relaxation measurements represented in Figure A 4 It also measures the voltage across the chip heaters which is mostly meant as a diagnosis tool Particular module used is also equipped with a Model 2001 SCAN 10 Ch Scanner Card which alloys the instrument to change between different channels and measure different signals 117 KEITHLEY Figure A 4 Keithley 2000 6 2 Digit Multimeter Keithley 2002 Multimeter used to measure the difference between the thermopile voltages of the two Xensor microcalorimetry chips represented in Figure A 5 Mostly meant as a diagnosis tool 777 n _ 000 00020 mUDC 3 899 oe une im T9 1 C asa POWER Coman maan nnm mm m nme Figure A 5 Keithley 2002 8 2 Digit High Performance Multimeter The above four modules are connected to a PC via a GPIB bus that enables for their remote control with the main interface software see section A 3 1 Model 430 Power Supply Programmer AMI field control module used to apply and remove magnetic fields up to 9 Tesla represented in Figure A 6 Capable of both setting magnetic field ramps and stable continuous fields 118 lii lt a SEE Bee SV Figure A 6 Model 430 Power Supply Programmer Model 4Q0625PS AMI power Supply power supply fo
20. P systems Results on both these systems were inconclusive as both of them do not seem to exist as a single phase Still the exploration of the several secondary phases found in our Mn Co Si P samples revealed the existence of the Mn Co o Si P system which displayed an extremely relevant inverse magnetocaloric effect A Si free Tc tuning attempt performed in the Fe Co P system revealed the existence of the Fe Co S P system which displays a unable Tc with S content as well as a novel transition of an uncertain nature at 370 K Chapter 7 explores the Fe rich side of the Mn Fe P Ge magnetocaloric system The transition temperature of this system is extremely easy to tune with careful manipulation of Fe and Ge content as well as stoichiometrical proportions which gives rise to the real possibility of lowering the Ge content in this family of compounds and thus make it economically viable for practical magnetocaloric applications While the Mn Fe P Ge system suffered a significant loss of transition sharpness with reducing Ge content the same was not verified in the Mn Fe 95 P Ge system and at the lowest possible Ge content which coincides with the lowest possible monetary cost Tc was found to be tunable below a temperature of 225 K with the magnetic entropy changes in this area oscillating between 3 and 3 5 J kg K for a field of 2 Tesla Novel and unexpected magnetic properties observed in this system suggest an excit
21. a deeper understanding on the nature of this transition fundamental to the work developed in Delft the objective of the current study is to re investigate the results presented in Figure 8 1 using the microcalorimetry setup described in Chapter 4 8 3 Experimental procedure The sample used for this study was the same Fe P sample produced by the drop synthesis technique as described by Carlsson et al 20 which was also used by Caron et al 10 in the acquisition of the data displayed in Figure 8 1 The used copper reference consisted of a small 99 9 pure spherical particle from Alfa Aesar The details of both these samples can be found in Table 8 1 107 Table 8 1 Details regarding the sample and reference used in the measurements for the current chapter Mass mg Error mg Grease mg Error mg Sample 0 21 0 02 Grease mass was too small to be Reference 0 23 0 02 registered by our balance none the less we should consider a maximum value of 0 02 mg the balance error The measurement procedure consisted of a continuous ramp with a heating rate of 0 2 K min with a constant applied magnetic field ranging from O T to 0 8 T with steps of 0 1 T 8 4 Results and discussion The results obtained from the above mentioned measurements revealed to be a challenge to interpret From O to 0 3 T the observed transition displayed the characteristics of a first order phase transition although a slight decrease in the specific
22. an uncertain danger for its long term use 1 3 3 Heusler alloys These compounds usually undergo a 1 order temperature induced structural transition between a high temperature cubic austenite and a low temperature tetragonal martensite phase This transition is often accompanied by a magnetic phase transition between ferromagnetism and paramagnetism 14 For example the Ni MnGa compound undergoes a structural transition around 220 K and a magnetic transition at around 376 K 22 These transitions are easily tuned by altering the mentioned stoichiometric proportions to the point of coupling both into a single large magneto structural one 23 25 Some of these alloys containing In or Sn demonstrate an inverse MCE explained in Chapter 6 Apart from the typical large thermal hysteresis which can be controlled by compositional tuning processing or pressure 14 these systems also present significant difficulties since they are usually not reliable for long term thermal cycling unless they are produced as single crystals 14 1 3 4 Fe P based compounds Fe P type compounds usually undergo a 1 order transition consisting of an elastic shift of c a with hardly any volume change accompanying a magnetic transition from a ferromagnetic to a paramagnetic phase By the substitution of As B and Si into the P site and the partial substitution of Fe by Mn it is possible to increase the transition temperature of this compound from 217 K 26 up
23. frame b detailed view on the calorimetry chip itself Chip specifications are displayed in Table 4 1 Table 4 1 XEN 39328 chip manufacturer s specifications at 22 C These chips are largely similar to the older model XEN 39287 used in the earlier stages of the development of the current setup Membrane dimensions 0 9x0 9 mm Approximate thermopile sensitivity 2 0 mV K Heating site dimensions hotspot 92x92 um Pins TO 5 Heater resistance 1 2 kQ Heater resistance temperature coefficient 0 1 Kt Effective heat capacity in air 100 nJ K Maximum heating voltage in vacuum 2 7 V Membrane thermal resistance 50 100 kK W Membrane thermal resistance O K temperature coefficient Thermopile resistance 30 kQ As is shown in Figure 4 2 in the current setup we use two of these chips one loaded with a Cu reference and the other with our sample In contrast with the single chip setups used by both Morrison et al 10 and Minakov et al 13 this two chip 38 setup was chosen in order to minimize any chip dependent issues that might interfere with our measurements Assuming that the two chips used are equal in all aspects and are under the same conditions we can thus use the heat capacity ratio between them and disregard most of the chips influence on the measured results making our subsequent data analysis simpler This approach eliminates the necessity of having to deal with the unknown Seebek coefficient of the chip
24. in fact exist as a single phase material The sought after hexagonal P6 mmm phase discovered for the Mn Fe Si P is not present in the Fe Co 3 Si P system 75 Fe rt 10 y 900 C 1173K Fe Co ht 10 20 30 40 50 60 70 80 90 Co Co P at P P Figure 6 2 Phase diagram of the Fe Co P system The circle demonstrates the compositional limit of Co substitution on the Fe P pure compound contradicting the results presented by Goto ef al 8 and Liu et al 9 Figure taken from 23 76 EG Normalized Specific Heat 200 300 400 500 600 700 Temperature K c o Current Work Fe CoP 4 Ref 9 9 12 Co limit according Co limit according as 7 to ref 8 and 9 to ref 21 S a 2 9 06 ng 9 dile i 9 00 g S 447 o S T 444 4 41 0 0 0 2 0 4 0 6 08 1 0 1 2 1 4 1 6 1 8 X Co content Figure 6 3 a Normalized DSC measurements for the samples Fe P Fe CoP and Fe 3Co P demonstrating the decrease in Tc with increasing Co content b Evolution of the lattice parameters a and c with increasing Co content in the samples produced for the current work and compared with the values given by Hui ping et al 9 6 3 3 Sulfur substitution in Fe Co 3 P Si Given the undesired results of Si substitution in the P rich side of the Fe Co 3 Si P system one final attempt was made to further tune down the extremely 77 high Tc of these
25. lowered and before you screw it shut repeat point 4 5 and 11 Screw it shut and attach the vacuum pump to the insert top Start pumping and gently open the vacuum valve so as not to create a pressure chock which might dislodge your chips or damage the setup After you have created some vacuum 10 or 10 repeat point 4 5 and 11 Make a test measurement of a few minutes to make sure everything is responding as it should see section A 5 Fill the liquid nitrogen tank and repeat point 4 5 and 11 Fill the liquid helium tank and repeat point 4 5 and 11 Yes Gently attach the vacuum pump to the insert top avoiding tilting it too much Start pumping and gently open the vacuum valve so as not to create a pressure chock that might dislodge your chips or damage the setup After you have created some vacuum 10 or 10 repeat point 4 5 and 11 Make a test measurement of a few minutes to make sure everything is responding as it should see section A 5 Open the cryostat and gently lower the insert into it screw it shut and repeat point 4 5 and 11 A 5 Measurement protocol What follows are step by step instructions for the performance of both temperature and magnetic field dependent specific heat measurement in the microcalorimetry instrument currently described At this point there will be no concern with data handling or calculation merely data acquisition Should you have performed all of the instructions given in section A 4 co
26. magnetic fields suggesting two magnetic solutions to be nearly degenerate in energy 15 Under high pressures 1 5 GPa this compound turns antiferromagentic 16 Senateur et al claim it actually undergoes a structural transition to the mm2 structure Figure 7 10 with the application of high temperatures and external pressure 8 GPa and 800 C 17 This compound displays a great structural sensitivity to alloying with a third element in terms of Tc magnetic moment tuning 15 18 and structural changes which frequently gives rise to the already mentioned orthorhombic mm2 structure and metamagnetic and antiferromagnetic states 15 18 Previous heat capacity measurements have shown that the already mentioned first order phase transition is the only observable magnetic contribution manifesting itself as a sharp and symmetric peak at T 19 Focusing on data for the polycrystalline samples displayed in the study by Caron et al 10 for fields below 0 05 T this transition presents the characteristics of a first order transition which rapidly changes into a broader second order transition when the external applied field is increased a phenomenon that is also accompanied by an increase in Tc as displayed in Figure 8 1 106 100 polycrystal 80 60 M Am kg 40 20 0 mm 100 150 200 250 300 350 400 T K Figure 8 1 Magnetization vs temperature at various fields for polycrystalline Fe P 10 In order to acquire
27. not able to do so in a way that is beneficial for future magnetocaloric application 66 5 666 5 664 z 5 662 Q 44 5 660 A e Transition temperature T K 5 658 o o 0 0 0 1 0 2 0 3 0 4 0 5 B content x Figure 5 11 Evolution of the lattice parameter a and Ty of MnFe SiB with increasing B Given the increasing presence of undesirable secondary phases in samples with B constants above 0 15 the lattice parameter values above this composition should be regarded as rough approximations 5 7 Conclusions Based on the collected data and previous literature the magnetostructural quaternary phase diagram of the Mn Fe 3 Si P system was successfully constructed offering a rare overall view of its unique magnetic and structural properties and greatly expanding our knowledge of the already widely studied Mn Fe Si and Mn Fe 3P systems Still this diagram appears to be very particular and sensitive to the sample production procedure as other authors have found different properties when using different annealing temperatures The novel hexagonal phase observed for the first time in this system like all other phases shows paramagnetic behaviors at low Fe concentrations and ferromagnetic at high concentrations and may offer the possibility for further developments on those areas where both the Mn Fe 3Si and Mn Fe 3P systems
28. o E o 400 L K s ai 1 200 0 10 y 20 0 10 X 20 Figure 5 1 a Saturation magnetization and b transition temperature variation with composition for the Fe Mn Si system The insert in a represents the Typical magnetization temperature curve in the range 0 7 x 1 8 In b Tr stands for the second transition between the antiferromagnetic and canted ferromagnetic phase 11 Regarding site preferences of the Fe and Mn atoms in the parent compound FesSi the addition of Mn atoms preferably occupies the A sites for compositions between O and 0 75 Within this interval Tc continuously decreases from 840 to 370 K while the magnetic moment of this site remains relatively constant meaning that Fe A and Mn A atoms exhibit approximately the same moment 13 For compositions above 0 75 the Mn atoms begin to occupy half of the B sites so that the B sublattice can be split into B and B sites 15 The A sites are only found to be completely filled for a Mn content above 1 5 13 For Mn composition between 0 75 and 1 8 we then find the already mentioned canted ferromagnetism displaying a transition between canted ferromagnetic and antiferromagnetic states 13 14 49 5 2 2 The Mna Fe P system As far as practical application goes this system has been studied in the context of nuclear reactor material research mostly the FesP compound 16 Beyond this it does not seem to provoke much interest outside
29. of a more efficient refrigeration system Contrarily to vapor compression this technology resorts to materials in solid form and does not use hazardous gases being able to reach a maximum theoretical efficiency of about 6096 3 being a bright promise for the future 1 2 Magnetocaloric effect In simple terms the Magnetocaloric Effect a phenomenon already widely known since the XIXth century but only named as such in 1917 1918 by P Weiss 5 is the increase in temperature of a magnetic material with the application of an external magnetic field This can be understood if we imagine a magnetic material with randomly aligned spins to which we apply an external magnetic field Considering an ideal system as we apply this field the spins in the material will tend to alight with it causing a reduction in the magnetic entropy of the system AS and generate a corresponding heat transfer 00 Both these quantities are related by the second law of thermodynamics for a reversible process as 22 1 1 The thermo magnetic cycle is in all aspects analogous to the vapor compression cycle of the current commercial refrigerators in which we merely substitute a gas by a magnetic material and pressure by an external magnetic field as exemplified by Figure 1 1 The field in question can either be applied under adiabatic or isothermal conditions corresponding to either an adiabatic temperature change where the temperature of the system material in
30. of this table was itself performed in the early stages of the setup assembly using a calibrated Cernox thermometer from LakeShore Cryogenics 36 1000 100 TL S 2 otc LG o 1 zn 10 o a 0 50 100 150 200 250 300 Temperature K Figure 4 3 Resistance vs temperature for the carbon glass temperature dependent resistor used to determine the temperature inside the sample space Furthermore this temperature controller determines and supplies current to the heater by comparing the measured temperature with a set point temperature As cooling is achieved passively by the simple contact of the insert with the LHe this heater is the only active element in the temperature control 4 2 3 Microcalorimetry chips As the use of strong magnetic fields in ferromagnetic samples can give rise to high field gradients that can affect the measurement by moving or dislocating samples we have decided to use small samples in the order of micrograms to minimize this effect For this end we used two XEN 39328 microcalorimetry chips manufactured by the company Xensor Integration These chips consist of a thin 0 9x0 9 mm SiN membrane 17 with a sensitive thermopile and a heater Given their small size these chips have a high thermopile and heater accuracy Detailed images of these chips can be observed in Figure 4 4a and 4 4b 37 Figure 4 4 XEN 39328 microcalorimetry chips used in the current setup a overall detailed view on the chip and
31. orbit lattice interaction meaning that the electronic orbitals are coupled to the crystal lattice of the system 4 This strongly influences the orientation of the moments with respect to the lattice which mathematically translates itself as different values for the above mentioned anisotropy constants This is referred to as magnetocrystalline anisotropy 2 2 2 2 Anisotropy energy density according to structure The anisotropy constants are not usually defined in theoretical terms but rather through measurement being in fact material specific quantities 4 The values of these constants have very specific effects on the energy surface defined by Equation 2 13 depending on the symmetry of the lattice 2 3 Uniaxial system Specifying the expansion of Equation 2 13 for a system with a single axis of high symmetry assumed to be the z axis the following expression is obtained 14 Fin n K sin 8 2 14 When K is positive the direction of lowest energy the easy axis is z When K is negative we instead have an easy plane perpendicular to z 4 Hexagonal Tetragonal and Rhombohedral systems For these three cases the anisotropy energy density has the following forms Hexagonal fay Gn K sin O K sin 0 K sin 0 cos6 2 15a Tetragonal 6 fay Gn K sin O K sin 0 K sin 0 sin 44 2 15b Rhombohedral 6 fan m K sin 0 K sin 0 K cos sin O cos3 2 15c Considering that
32. phase however is no longer observed at an Fe substitution of only 0 5 Instead the hexagonal structure consistent with the P6 mmm space group is detected This structure is only present in the P interval between 0 2 and 0 9 bordered by the cubic phase of the Si rich side and by the tetragonal phase on the P rich side It also exhibits antiferromagnetic behavior up to the Fe content of 1 2 The Ty of both the hexagonal and the tetragonal structures also demonstrates a slow increase with Fe content At an Fe content of 1 on the P rich side of the diagram the orthorhombic structure is observed This structure appears to be consistently antiferromagnetic with a relatively constant Ty of 340 K and is observed as far as the composition Mn 9sFe 05S10 5Po s At an Fe substitution of 1 2 both the cubic and hexagonal phases start to display ferromagnetic behavior The Tc of both these phases display an increase with Fe content although the cubic phase does so at a much higher rate From a Fe content of 1 2 up to 2 3 the cubic phase besides the transition between ferromagnetic and paramagnetic phases also displays a transition between antiferromagentic and ferromagnetic phases with a relatively constant Ty of 50 K This behavior is described as canted ferromagnetism At an Fe content of 1 9 the hexagonal phase is no longer observed being substituted by the cubic phase whose range can now be observed from 0 to 0 5 P substitution From the Fe cont
33. rare earth transition metal magnets on which we can group Sm Fe Nd Fe B and Sm Co magnets 18 20 22 24 Given the current demand for high performance magnets the most widely used are the rare earth transition metal compounds particularly the Nd Fe B as can be observed in Figure 7 9 These are relatively expensive materials as a consequence of the limited yearly rare earth production rate 19 25 Such a situation logically makes the research on novel permanent magnet materials that do not contain rare earth elements an exciting and extremely relevant field of investigation 97 B Nd Fe B B Sm Co Alnico B Ferrite Figure 7 9 Breakdown of the permanent magnet world market 19 The magnetization versus temperature in field cooled FC and zero field cooled ZFC conditions for the Mno Fe s Poo9Geo sample show a striking difference in magnetic behavior The FC curve displays a considerably higher magnetization than the ZFC curve Given that this sample consisted of an irregular multigrain powder such a behavior may indicate the presence of strong magnetic anisotropy in this compound This observation suggests that there is an easy direction in this compound along which the magnetic moments will preferably align an essential property of a permanent magnet As the relevant intrinsic properties that should be displayed by a material to make it relevant for permanent magnet applications can be listed as 1 a high saturatio
34. sites coplanar with the P and Ge atoms at the 2c site 11 For Fe Mn P Ge the Mn and Fe atoms occupy separate planes as depicted in Figure 7 2 2c site 3f site P Ge 2 Fe Mn C X SMF geste lb site NPyramid N 7 Mn Fe_ P Ge l Fe Mn P Tetrahedron Figure 7 2 Crystal structure of the Mn Fe P Ge system presenting its several atomic sites the arrows represent the magnetic moments of Mn and Fe atoms while the system is in its Ferromagnetic state 11 This system is paramagnetic at high temperatures and ferromagnetic at low temperatures 11 Its 7c is easily tunable over a broad temperature range 150 K Tc 450 K by careful adjustment of both Mn Fe and P Ge ratios 1 Accompanying the magnetic transition at Tc this system also undergoes a contraction of the lattice parameter c and an expansion of a 11 which can be identified in X ray diffraction data as a shift in both the 300 and the 002 peak 89 towards higher and lower angles respectively This phenomenon is illustrated in Figure 1 3 500 1 00 0 80 450 0 60 400 0 40 x 0 20 D 350 5 0 s 300 E 250 O H 200 150 51 52 53 54 55 56 20 deg Figure 7 3 Contour plot of the angle change in the 300 211 and 002 peaks versus temperature for Mn FeosP 75G 0 25 ilustrating the sharp discontinuity in lattice paramenters acompaning the magnetic transition 12 7 3 Sample preparation and char
35. temperature for the Mno Fe1 4Po sGeo2 sample demonstrating the slow and smooth character of the magneto elastic transition The transtion temperature of this sample has been determined as being 358 K by DSC measurments The bar on the right of the figure represents the normalized peak intensity 7 4 2 Mn Fe 95 P Ge system In contrast to the Mn Fe P Ge system the Mn Fe os P Ge system does not seem to be so negatively influenced by the increase of Fe content as it maintains a usable sharp transition up to high Fe concentrations This stability has enabled us to reduce the Ge content down to Ge 0 12 less than half of the value used by Trung et al 1 Lowering the Ge concentration beyond 94 this point was found to trigger antiferromagnetic behavior in this system This concentration is therefore considered as the lowest possible Ge content in which this system still displays a sharp Ferro Paramagnetic transition as indicated in Figure 7 7 120 Mn Fe P Ge i 1 43 0 88 110 4 Mn Fe P Ge 1 43 0 9 100 Magnetization Am kg o e 0 50 100 150 200 250 300 350 400 Temperature K Figure 7 7 Comparison between the Ge 0 12 and Ge 0 1 magnetization versus temperature measurements in the Mn Fe os P Ge system Having determined a usable minimum Ge concentration we were able to tune Tc by changing Fe content in accordance with the behavior observed for the Mn Fe P Ge system displ
36. the academic sphere Both the Mn3P and Fe P compounds crystallize in the tetragonal 14 structure 17 18 Ni P prototype Figure 5 2b with three different Mn Fe sites 18 Mn3P is an antiferromagnet with a Ty of about 115 K and a magnetic moment of 1 69 u per Mn atom 19 FesP on the other hand is a ferromagnet with a Tc of about 700 K and a magnetic moment of 1 89 ug per Fe atom 19 Mn Fe atoms Si P atoms Figure 5 1 a The Mn Si Fm3m cubic structure showing the Mnl atoms on the A sites the MnII atoms on the B B sites and the Si atoms on the C sites b The MnjP and Fe P I 4 tetragonal structure The Mn Fe P system maintains the same 14 tetragonal crystal structure for O lt x lt 1 20 21 and again for 2 2 lt x lt 3 19 21 approximately In the gap between these two intervals an orthorhombic structure is reported 21 In this structure the moments order antiferromagnetic with a relatively constant Ty of about 340 K 21 Goto et al 21 having studied the magnetic properties of Fe M P with M Cr Mn Co Ni and report that this orthorhombic structure only occurred in the Mn 50 compounds all the other systems maintained the I 4 tetragonal structure for the studied compositions 5 3 Sample preparation and characterization All produced samples belonging to the Mn Fe Si P system were prepared from the appropriate amounts of 99 iron powder 99 99 gran
37. this depending on what method you used to perform your temperature ramp you may or may not need to do this step particularly if you used Zone settings 155 Finally remove the applied magnetic field following the instructions given in the Model 430 Power Supply Programmer User s Manual A 6 Data management What follows are step by step instruction for the handling of the data file produced during a specific heat measurement in the current microcalorimetry instrument 0 To prevent data loss or confusion it is advisable to move the produced meas tdms and meas tdms index files to a new folder specifically meant for working on this measurement 1 Open the Calorimeter Read vi 2 Turn this program on using the LabView On button and opening the Input file Path selecting the meas tdms file relative to the measurement you wish to work on 3 Press the List button on the bottom of the software panel This should list all the information recorded in your measurement file in the two columns entitled Channel X and Channel Y according to module and data type as shown in Figure A 33 156 Input File Path Save File b D SharedData calorimeter Peter Jose Fe2P 16 11 12 meas tdms 0 00 L 1 10 00 00 15 00 00 1 05 00 00 D SharedData calorimeter Peter Jose test measurments test tdms Channels x 2000 dme 2000 channel Channels y BAO me 2000 chan
38. values of the anisotropy constants Gd Co b c plane 90 180 270 q degrees Figure 2 6 Torque measurement for the b c plane in Gd Co 11 Table 2 3 shows examples of the room temperature anisotropy constant values for several magnetic materials As usually lower symmetry leads to larger anisotropy materials with a hexagonal crystal structure have larger anisotropy constants than those with cubic crystal structures 21 Table 2 3 Examples of anisotropy constant values at room temperature 4 Structure Compound K 10 J m K 10 J m Cubic Fe 4 8 0 5 Ni 0 5 0 2 FeO 1 1 MnO 0 3 NiO 0 62 MgO 0 25 CoO 20 Hexagonal Co 45 15 BaO 33 YCos 550 MnBi 89 27 References 1 2 3 4 5 6 7 8 9 22 A M Tishin and Y I Spichkin The Magnetocaloric Effect and its Applications Inst of Phys Publishing Bristol UK 2003 G Bertotti Hysteresis in Magnetism for Physicists Material Scientists and Engineers Academic Press San Diego USA 1998 B D Cullity and C D Graham ntroduction to Magnetic Materials Wiley IEEE Press Hoboken New Jersey 2009 K H J Buschow and F R de Boer Physics of Magnetism and Magnetic Materials Kluwer Academic Publishers New York USA 2004 Joong Jung Kim Hyun Soon Park Daisuke Shindo Satoshi Hirosawa and Hideyuki Morimoto Mater Trans 47 2006 907 L D Landau
39. wasn t for the massive load of work you took on Even if I was your supervisor I am humble enough to say that I owe you Logically now I must pass a review of my PhD and Pos doc colleagues both present and absent Swapna you are a trustworthy and honest person if there ever was one sometimes even too much with you we always know were we stand and don t let anyone ever tell you this isn t a virtue Anca you re too sweet and I m happy I got to see some of that sweetness come into the world as a whole new person Markus you re my friend that has been settled but do know that you were very close to being punched in the face a number of times Anna relax leaving your desk to have lunch with your friends isn t a sin Shasha you are the only girl I know who can successfully wear a leopard pattern shirt and if for nothing else since you came into the office at least somebody would always greet me with an enthusiastic Good Morning everyday you also liked my Bacalhau Br s so that s another virtue right there In s with you around there was a clear upgrade in conversation I could finally curse and swear and somebody would understand Luana doesn t count Fengjiao you re a ray of sunshine Romain I got the feeling you went through some rough times during your PhD but I m happy to see that you came out on top Gijs it was good to know somebody else who remembers the Muppets and the movie Krull With all of you lunch time was a
40. 0 time F 2400_time 2400 volt 2400 volt 2400 curr 2400 curr 331 time 331 time 331 setpoint 33 1 setpoint 331 temp 33 1 temp 33 1 result 331 result 331 heater 33 1 heater ay m X Axis Y Axis 2000 time 2000 volt ave stop X axis Format List Read OK sof Figure A 35 Location of the functions mentioned in the above instruction for the Calorimeter Read vi The default names for these data sets which will be used from this point on are the following 1 2000 time vs 2000 volt channel csv 2 2400 time vs 2400 volt switch cvs 3 33 time vs 331 temp temperature csv 6 Open the channel csv do not use MS Excel Analyze the first data points in this file and delete those that clearly do not make sense as shown in Figure A 36 160 nes en ape e ee EA 0 046878 0 105015 0 093755 0 094663 0 140633 0 073961 0 187511 0 052993 0 234388 0 031906 0 281266 0 021311 0 343770 0 000011 0 390647 0 000012 0 437525 0 000001 0 484403 0 000000 0 531281 0 000011 0 593784 0 000014 0 640662 0 000002 0 687540 0 000001 734417 0 000012 81295 0 000014 3799 0 000002 76 0 000000 554 0 000011 0 984432 0 000025 a enbi 1 093813 0 000023 1 140691 0 000035 1 i la o e to MD D y s Ww WO O e m o wo 187568 0 000036 34446 0 000024 81324 0 000022 2 2 be deleted in the channel csv file Figure A 36 Initial data points that should 7 Open Ma
41. 04 5 0 05 0 06 0 07 0 08 0 08 0 10 2 11 8 Auto 00 aj x 1 1 1 1 1 05 00 00 10 00 00 15 00 00 20 00 00 00 00 Axis 6 E X axis Format Tl gea 9191015 11 1 Figure A 22 Front panel of the data visualization software Table A 8 Legend for Figure A 22 Input file box This refers to the original tdms file produced by the Calorimetry vi the data acquisition software 2 Output file box This refers to the path name and file type you produce with this software 3 Data visualization panel This displays the two data sets you define as X and Y in points 5 and 6 4 Visualization tools 5 Double listing of all the data contained in the tdms file produced by the Calorimetry vi 6 Data selected from the double listing in point 5 as X and Y 7 Lists all the data contained in the tdms file produced by the Calorimetry vi selected in point 1 on point 5 8 Plots the data selected as the X and Y axis in point 6 into the visualization panel 9 Saves the data displayed in the visualization panel in a two column file as specified by the path name and file type in point 2 10 Global stop button This shuts down the whole data extraction software 11 X axis format This switches between the definition Time and Decimal allowing for the selection of the appropriate data format on the visualization
42. 1 Introduction 8 3 Experimental procedure 8 4 Results and discussion 8 5 Conclusions References setup described in Chapter 4 115 Al Hardware Overview sn tn terende 115 AD Wiring nn 120 A 2 1 Insert head 120 A 2 2 Insert body andtop 11 s 125 A 2 3 Insert cables 125 A 3 Software overview 126 A 3 1 Interface software 126 A 3 2 Data visualization and extraction software 22222222222 133 A 33 Calculation software eee 135 A 3 4 Data display assessment 140 A 4Measurement preparation protocol AD A 5 Measurement protocol 146 A 5 1 Software handling 146 A 5 2 Temperature sweeps s 152 A 5 3 Magnetic fields 000 s 154 A 5 4 Stopping a measurement R kC C O Oa L 155 A 6 Data management 190 OUM WM o ihe a a teenie DA pe PRA o GD a O octet DI CAD o fonts 165 Oameni caos caos ops caca a A A AS A ACAO fd 169 Acknowledgements LI LLLLLLLLL 173 List of Publication esesesese 179 Corriculum Vite 1 LLL 183 Epilogue ss 185 viii Chapter 1 Introduction 1 1 Global refrigeration Whether it is for food conservation or simple comfort the possibility to control environmental temperature is a general need in modern society Of these two points food conservation is without a doubt the most important one It is obvious that our modern life style is impossible to conceive without refrigerators and food preservation Historically refrigerato
43. 15 14 On the other hand positive pressure compression of the unit cell via hydrostatic pressure increases the AS value but decreases the transition temperature 16 Furthermore one can also substitute Fe by other elements such as in the La Fe Co 4 Al and La Fe Co 4 Si systems to increase the magnetic ordering temperature and have a significant MCE near room temperature 17 18 The tuning and control mechanisms of this system are somewhat elaborate having even the necessity of the introduction of a porous architecture in order to control its mechanical properties 19 and the use of La makes it a relatively expensive candidate as a working material for a commercial MCE refrigerator 1 3 2 MnAs based compounds The pure MnAs compound displays two different crystal structures a hexagonal NiAs structure at both low and high temperatures and an orthorhombic MnAs structure in the temperature range between 307 and 393 K The high temperature transition between these structures is 2 order while the low temperature one between ferromagnetic and paramagnetic states is a sharp 1 order magneto structural transition 20 displaying a significant thermal hysteresis 14 This compound can be effectively tuned by the careful substitution of As by Sb giving the possibility to shift Tc from 220 to 320 K while maintaining a large entropy change 21 The downside to this system is the use of toxic As in its composition which offers
44. 17 96 Although these results may seem less ideal than the previously mentioned ones from Trung et al 1 they represent a trade off between cheaper samples and improved mechanical properties which nonetheless still possess many possible applications in low temperature applications or in magnetocaloric material cascading in practical cooling devices 7 5 Permanent magnet potential Surprisingly during the course of this study certain limited compositional areas of Mn and Ge poor Mn Fe os P Ge have been found to possess very exciting magnetic properties which may indicate a definite potential for future permanent magnet applications The applicability for permanent magnets in modern society are widespread These are used in a massive array of modern appliances from the mundane to the cutting edge in science and industry everything from motors and generators dynamos microwave generators loudspeakers communications and electronics aeronautics electric cars nuclear industry satellite electric power systems measuring instruments and sensors medical devices MRI latching and holding devices separation equipment such as water and oil purifiers and so on 18 21 Among the families of materials used for such application we can cite the Alnico magnets Al Ni Co 18 21 22 the Ba or Sr hexaferrites characterized by lower flux densities and higher coercivities than the Alnicos 21 and as the most modern and widely used the
45. 3302 N K Sun W B Cui D Li D Y Geng F Yang and Z D Zhang Appl Phys Lett 92 2008 072504 P J Webster K R A Ziebeck S L Town and M S Peak Philos Mag B 49 1984 295 F Albertini F Canepa S Cirafici E Franceschi M Napoletano A Paoluzi L Pareti and M Solzi J Magn Magn Mat 272 276 2003 2111 V Khovailo V Novosad T Takagi D Filippov R Levitin and A Vasil ev Phys Rev B 70 2004 174413 A Vasil ev A Bozhko V Khovailo I Dikshtein V Shavron V Buchelnikov M Matsumoto S Suzuki T Takagi and J Tani Phys Rev B 59 1999 1113 O Beckman L Lundgren P Nordblad P Svedlindh A T rne Y Andersson and S Rundqvist Phys Scripta 25 1982 679 Chapter 2 Theoretical approach 2 1 Basic magnetocaloric quantities The isothermal entropy change AS and adiabatic temperature change AT quantities discussed in Chapter 1 can be deduced using the following mathematical steps Considering the Gibbs free energy G as a function of temperature T pressure P and Magnetization M used in an isobaric system we can write G U TS PV MH u 2 1 where S is the entropy and U the internal energy The total differential then becomes dG VdP SdT MdH 11 2 2 Yielding thus the expressions AG S T H P 2 3 en 5 Ce 1 AG M T H P 2 3b T H P zo 2 3b AG ET 2 V T H P e 2 3c Through mathematical development o
46. 4 Fm3m P6 mmm Fm3m P6 mmm Fm3m P6 mmm Fm3m Fm3m Fm3m P6 mmm Fm3m P6 mmm Fm3m Pmmm Magnetic Transitions Antiferro Para Antiferro Para Antiferro Para Antiferro Ferro Para Antiferro Ferro Para Antiferro Ferro Para Antiferro Ferro Para Antiferro Ferro Para Ferro Para Ferro Para Ferro Para Antiferro Para Antiferro Para Antiferro Para Antiferro Ferro Para Antiferro Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Trans Temp K 27 24 5 50 62 137 50 233 401281 301342 121387 397 547 840 79 82 90 58 135 50 173 657 750 102 111 MnFe Sio Poa Mn3Sio sPo s Mn sFeo sSio sPo s Mn FeSio sPo s Mn Fe sSio sPo s Mn 4Fe 6Sio sPo s Mn Fe Sios Po Mn Fe 8Sio sPo s Mn Fe 9S1o sPo s MnFe Sio sPo 5 Mna abe 1Sio sPo 5 Mno Fe gt 2S1o sPo s Mna spe sSio sPo s Fe3S10 5Po s Mn Fe 8Si0 4P0 6 Mn Fe 9Sio 4Po s MnFe gt Sio4Po s Mn gFe 4Sio 3Po 7 Mn sFe sSio 3Po 7 Mn Fe 7Sio 3Po 7 Mn Fe Sio 3Po MnsSio Pos Mn Fe 5S10 2Pos MnFeSio gt Po s Mn Fe 5Sio Pos MnFe Sio 2Po s Mna abe 1Sio 2Po s Mn sFe gt gt Sio 2Po s Mn Fe 3Sio 2Po s MnyoFe 4Sio 2Po s Mn se 5Sio 2Po s Fe3Sio 2Po s Fm3m P6 mmm Pmmm Ferro Para I4 P6 mmm P6 mmm P6 mmm P6 mmm P6 mmm P6 mmm Pmmm P6 mmm Pmmm Fm3m Fm3m P6 mmm Pmmm Fm3m Pmmm P6 mmm Fm3m Pmmm Fm3mH 4 Fm3m 4 P6 mmm Pmmm P6 mmm Pmmm Pmmm Fm3m P6 mmm Pmmm P6
47. 4 232 240 248 Temperature K Figure A 38 Isolated specific heat measurement for polycrystalline Fe P under an external magnetic field of 0 3 Tesla 163 164 Summary Given the potentiality of Fe P based alloys for magnetocaloric application this thesis covers various offshoot material systems related to this rich family of compounds such as the Mn Fe 3 Si P Mn Co 3 Si P and the Fe Co 3 Si P Also covered is the assembly of a microcalorimetery setup meant for the study of certain fundamental qualities of these systems as well as an economical optimization of the Mn Fe P Ge system After a general and contextualizing introduction in Chapter 1 it proceeds in Chapter 2 with the coverage of theoretical aspects of the magnetocaloric effect as well as the description of magnetic anisotropy phenomenon both concepts being fundamental for the understanding of the results presented in the subsequent chapters Experimental procedures for sample preparation and characterization are detailed in Chapter 3 with the details of the experimental microcalorimetry setup being offered in Chapter 4 This specific heat measuring instrument with the capacity for the application of magnetic fields up to 9 Tesla resorts to microcalorimetry chips from the company Xensor Integration With this instrument is it possible to perform specific heat measurements with applied magnetic fields up to 9 Tesla in milligram samples This offers the possibility
48. 6 2005 043906 A A Minakov A W van Herwaarden W Wien A Wurm C Schick Thermochim Acta 461 2007 96 S A Adamovsky A A Minakov C Schick Thermochim Acta 402 2003 55 Mikhail Merzlyakov Thermochim Acta 403 2003 65 Sander van Herwaarden Gas Nanocalorimeters Xensor Integration 2010 J G Bunting T Ashworth and H Steeple Cryogenics 9 1969 385 W Schnelle J Engelhardt and E Gmelin Cryogenics 39 1999 271 M Wun and N E Phillips Cryogenics 15 1975 36 H J Schink and H v Lohneysen Cryogenics 21 1981 591 Nhu Hoa Kim Ngan Magnetic Phase Transitions in NdMn and Related Compounds Universiteit van Amsterdam 1993 PhD thesis Chapter 4 N Dass J Phys Soc Japan 28 1970 251 45 46 Chapter 5 Magnetostructural study of the Mn Fe P Si system 5 1 Introduction Previous material studies performed on the Mn Fe P Si and Mn Fe 95 P S1 systems 1 2 have demonstrated a clear experimental difficulty in achieving pure single phase samples in this family of compounds In the above mentioned references a residual amount of a cubic Mn Fe Si phase was consistently observed in samples belonging to these two systems The motivation of the research presented in the current chapter was to explore the magnetic behavior of this secondary phase generalized as the Mn Fe P Si system and evaluate its magnetocaloric potential This study has made it possible to outline the basic
49. A Smith Eur Phys J H 38 2013 507 6 P Debye Ann Phys 81 1926 1154 7 W F J Giauque Am Chem Soc 49 1927 1864 8 W F Giauque and D P MacDougall Phys Rev 43 1933 768 9 G V Brown J Appl Phys 47 1976 3673 10 V K Pecharsky and K A Gschneidner Phys Rev Lett 78 1997 4494 11 W Choe V K Pecharsky A O Pecharsky K A Gschneidner Jr V G Young Jr and G J Miller Phys Rev Lett 84 2000 4617 12 V K Pecharsky A P Holm K A Gschneidner Jr and R Rink Phys Rev Lett 91 2003 197204 13 L Morellon Z Arnold C Magen C Ritter O Prokhnenko Y Skorokhod P A Algarabel M R Ibarra and J Kamarad Phys Rev Lett 93 2004 137201 14 O Gutfleisch M A Willard E Br ck C H Chen S G Sankar J P Liu Adv Mater 23 2011 821 15 16 17 18 19 20 21 22 23 24 25 26 A Fujita S Fujieda Y Hasegawa and K Fukamichi Phys Rev B 67 2003 104416 A Fujita S Fujieda K Fukamichi H Mitamura and T Goto Phys Rev B 65 2001 014410 F X Hu B G Shen J R Sun and Z H Cheng Phys Rev B 64 2001 012409 J R Proveti E C Passamani C Larica A M Gomes A Y Takeuchi and A Massioli J Phys D Appl Phys 38 2005 1531 J Lyubina R Sch fer N Martin L Schultz and O Gutfleisch Adv Mater 22 2010 3735 H Wada and Y Tanabe Appl Phys Lett 79 2001
50. A 11 Chip terminals legend TP TP Positive negative poles of the chip thermopile source of calorimetric data RHI RHI Positive negative poles of the chip heater current RH2 RH2 Positive negative poles of the chip heater voltage Sample space wire legend Wires 3and 4 Wires connecting the chip sockets to the Keithley 2400 Source Meter on the outside of the setup the source of the square wave which is fed into the chip heaters and consequently heats up the samples in the chips enabling for the performance of relaxation measurements 122 Table A 1 cont Legend for Figure A 11 Wire 1 Middle point between the right RH1 and the left RHI It does noting in the current setup and it is not connected to any equipment on the outside of the cryostat it is used merely for diagnosis purposes Wires 10 12 Wires connecting RH2 and RH2 of the left and right chips respectively 9 and 11 to the Keithley 2000 Multimeter on the outside of the setup making up channels 1 and 2 in this equipment respectively Wires 15 14 Wires connecting the calorimetry chip s thermopiles to the Keithley 2000 and 13 Multimeter on the outside of the setup making up channels 3 and 4 respectively Theses are the channels carrying the information of the thermal response of the samples mounted on the two chips to the heat pulses provided by the chip heaters making them the sources of measured
51. Co Si compound From this Co content on an increase in lattice parameters is observed 7 with the Co FeSi compound having a lattice parameter of 5 658 A 4 In terms of magnetic behavior the Si rich side displays particularly high values of Tc with 1100 K being reported for the FeCo Si compound 1 3 The saturation magnetization is also reported to increase with Co addition 4 5 On the P rich side of this system there seems to be little information available Several studies claim that there is a clear compositional limit of about 30 to the amount of Co that can be effectively substituted in pure FesP 8 9 and therefore that the Co3P compound does not exist 9 10 Below this Co compositional limit this system presents the already studied 14 tetragonal structure the orthorhombic phase observed in the Mn Fe P system is not present for the Fe Co 3 Si P system 8 10 In the 7 4 structure both a and c parameters decrease with Co addition 8 The same is true for both Tc and the overall saturation magnetization 8 10 6 2 2 The Mn Co Si P system The Si rich side of this system is once again a relevant study object for spintronics 11 14 and electronic conductivity 15 displaying a high Curie 12 temperature 12 13 as most metallic compounds containing Co and a large gap of 0 4 eV in the minority spin channel being a system frequently used for the fabrication of magnetic tunnel junction 11 12 14 The ful
52. Co powder from Alfa Aesar 6 2 Material overview 6 2 1 The Fe Co 3 Si P system On the Si rich side the Fe Co 3 Si P system has been mostly studied in the field of spintronics 1 with particular emphasis on spin related transport properties 2 similarly to the Mn Fe 3Si system Co based Heusler alloys particularly the FeCo5Si composition exhibit remarkably high magnetic moment and Tc 3 and have been studied in lateral spin valve devices for their giant spin signals at room temperature 2 This system shows a unique preferential site substitution pattern that allows a direct study of the correlation between the magnetic field interactions and local electronic configurations 4 5 Similarly to the Mn Fe 3Si system the Fe Co Si is reported to crystallize in the cubic Fm3m structure 1 3 4 Contrarily to the Mn Fe system this structure is not found to be stable throughout the whole compositional range between Co3Si and FesSi Those samples with Co content above 2 15 are not single phase 5 This crystal arrangement has the particularity that those atoms found on the left of Fe in the periodic table preferentially enter into the A site of its lattice while those to the right and beneath Fe the current case of Co preferentially occupy the B and B sites 4 6 see Figure 5 2a Starting from pure Fe3Si the lattice parameters originally decrease with increasing Co content dropping from 5 65 A to below 5 64 A for the Fe
53. Fe 3 P Si system Journal of Alloys and Compounds 520 2012 52 2 J V Leit o P van Dommelen F Naastepad E Br ck Description of a Differential Setup For Relaxation Microcalorimetry 5th International Conference on Magnetic Refrigeration at Room Temperature Proceedings 2012 263 3 J V Leitio M van der Haar A Lefering E Br ck Magnetic and Magnetocaloric Exploration of Fe rich Mn Fe P Ge Journal of Magnetism and Magnetic Materials 334 2013 49 4 J V Leit o P van Dommelen F Naastepad E Brick Description of a Differential Setup For Relaxation Microcalorimetry International Journal of Refrigeration 2013 Special issue based on selected articles presented on the Thermag V Accepted and in press Others X ray diffraction patterns accepted for the International Center for Diffraction Data ICDD 15 of November 2012 1 Fe P 2 Fe3Si 3 Mn P 4 MnsSio Pos 5 Mna sFe2 5Si 6 MnogFe 4Si 7 Mno Fe 3Si 8 Mno gFe 281 9 MnooFe Si 10 Mn 5Feo sP 11 MnFesSi Non physics related publications Published before the current PhD 1 Jos Leit o Um Argumento pela ideia de P tria Nova guia 1 2008 37 2 Jos Leit o Vadios e Poetas Uma considera o sobre a m stica Agostiniana Nova guia 3 2009 35 180 Published during the current PhD 1 Jos Leit o Ipsissimus Pessoa Nova guia 7 2011 99 2 Jos Leit o On Pessoa A H
54. L P Pitaevskii and E M Lifshitz Electrodynamics of Continuous Media Volume 8 Butterworth Heinemann Burlington USA 1984 K J Strnat in E P Wohlfarth and K H J Bushow Eds Ferromagnetic Materials Volume 2 Elsevier Science Publishers B V Amsterdam Netherlands 1980 W Sucksmith and J E Thompson Proc Roy Soc A225 1954 362 V S Ram and P Gaunt J Appl Phys 54 1983 2872 10 11 K D Durst and H Kronm ller J Magn Magn Mater 59 1986 86 J J M Franse S Sinnema R Verhoef R J Radwanski F R de Boer A Menovsky CEAM Report Elsevier Science Publishers B V London UK 1989 23 24 Chapter 3 Experimental equipment 3 1 Introduction This thesis is largely the result of experimental work of production and characterization of metallic samples In this chapter all experimental equipment used in this process is listed and detailed All materials described in the next chapters were produced at the section of Fundamental Aspects of Materials and Energy FAME of the Faculty of Applied Sciences of the Delft University of Technology The subsequent characterization measurements performed on these were equally performed in this same section unless mentioned otherwise 3 2 High energy planetary ball mill High energy ball milling has received great attention for its great range of applications from cold welding grain size control synthesis of equilibrium and non equi
55. O ao L S cart log data abs dlmread pif filename find pif 2 l end data abs 2 data length lt change logarithm CSV Seperator 2 lt prritliliend ly 211 find indices where heater voltage goes down ifstop find pif 2 1 end 2 gt pif 1 1 end 1 ndices where heater voltage goes up gt pifstart 1 S 2 Sfind ifstop pifstop find pifstop Sskip irst if stop comes before start ifstarttime Pik pirstart 2 1 1 D O mo Ho ifstoptime pif pifstop jo gt Find step down curves startdownwards find data delta delta bar data length downwards 1 for index 1 min size pifstarttime size pifstoptime 1 start find data time gt pifstarttime index 1 first stop start 5000 if length start 1 amp amp length stop 1 amp amp stop gt start 1 indices at which the downward curve starts and stops downwards downwards start stop end end 9 Find fit parameters from curves resultdown linlogmax 1 error 137 high 14 for index 1 size downwards timedown data_time downwards index 1 nSkip downwards index 1 nLength logdown data_log downwards index 1 nSkip downwards index 1 nLengtn P polyfit timedown logdown fitorder resultdown resultdown P fit parameters if use pif as switching time
56. Thermomagnetic studies on transition metal pnictides Proefschrift ter verkrijging van de graad van doctor aan de Technische Universiteit Delft op gezag van de Rector Magnificus prof ir K C A M Luyben voorzitter van het College voor Promoties in het openbaar te verdedigen op vrijdag 17 januari 2014 om 10 00 uur door Jos Carlos VIEIRA LEIT O Mestre em F sica Universidade de Aveiro geboren te Leiria Portugal Dit proefschrift is goedgekeurd door de promotor Prof dr E Br ck Copromotor Dr ir N H van Dijk Samenstelling promotiecommissie Rector Magnificus voorzitter Prof dr E H Br ck Technische Universiteit Delft promotor Dr ir N H van Dijk Technische Universiteit Delft copromotor Prof dr R Burriel Lahoz University of Zaragoza Spain Prof dr V B Sequeria Amaral University of Aveiro Portugal Prof dr R A de Groot Radboud Universiteit Nijmegen Prof dr A Schmidt Ott Technische Universiteit Delft Prof dr K H J Buschow Universiteit van Amsterdam Prof dr C Pappas Technische Universiteit Delft reservelid The work presented in this PhD thesis is financially supported by BASF New Business and Stichting voor Fundamenteel Onderzoekder Materie FOM under the Industrial Partnership Programme IPP I18 of the Stichting voor Fundamenteel Onderzoek der Materie FOM which is financially supported by the Nederlandse Organisatie voor Wetenschappelijk Onderzoek NWO Casimir PhD series De
57. about 25 K being paramagnetic above this temperature 3 5 7 Fe3S1 crystallizes in the same cubic Fm3m structure as Mn3Si The Fe atoms on the A site have a magnetic moment of 2 4 ug while those on the B B site exhibit a magnetic moment of 1 2 ug 8 9 The moments are ordered ferromagnetically at room temperature with a Tc of about 840 K 10 16 The whole pseudo binary Mn Fe Si system maintains the same stable Fm3m crystal structure for O lt x lt 3 A minor secondary phase either tetragonal or hexagonal has been reported by S Yoon et al for Mn contents between 1 8 and 3 12 Starting from the Fe4Si compound with increasing Mn content Tc drops linearly from 840 to approximately 300 K at a composition Mn Fe 381 16 12 From this composition on a transition between ferromagnetic and antiferromagnetic order is found around 50 K 13 This ferromagnetic state was described by Miki et al as canted ferromagnetism 14 This second transition eventually fades away at a composition of Mn sFe 5Si In this interval Tc decreases at a much slower rate reaching a value of 65 K at the above mentioned composition as can be observed in Figure 5 1 Further addition of Mn from this point on triggers an antiferromagnetic behavior in this system Ty drops smoothly from the last value of Tc until it reaches 25 K at Mn Si 14 48 a ag b E S os nal S Or L Oo E oO 9 TA O 4 o i t 800 b X G Ww TS 5 3 500 q c C
58. acterization All Mn Fe P Ge and Mn Fe os P Ge samples were prepared from the appropriate amounts of 99 iron powder 99 red phosphorous powder 99 5 binary Fe P powder 99 999 Ge chips and 99 9 manganese chips reduced at 600 C under a hydrogen atmosphere in order to remove oxides The samples where ball milled as described in Chapter 3 then compacted into 10 mm pellets with a pressure of 150 kgf cm and sealed into quartz tubes with an atmosphere of 200 mbar of argon Finally these were annealed in a vertical resistive furnace at 1100 C for 10 hours homogenized at 1000 C for 60 hours and then quenched to room temperature in water 90 This particular heat treatment was selected based on the study performed by N T Trung on the influence of annealing procedures in the Mn Fe P Ge system that indicates that this heat treatment produced samples with the sharpest magnetic transitions 13 The magnetic and DSC measurements were performed on all samples and homogeneity and crystal structure were checked using X ray diffraction as described in Chapter 3 7 4 Result and discussion 7 4 1 Mn Fe P Ge system Throughout the studied compositional range all samples belonging to both the Mn Fe P Ge and the Mn Fe os P Ge systems presented the Fe P type hexagonal structure with the occasional occurrence of a minor MnO peak in their X ray diffraction patterns The lattice parameter a was found to increase with Ge addi
59. ado onde cheguei sem a tua ajuda e o cornetim ainda n o se ouviu na pra a O guitarrista Nuno aprender guitarra consigo foi por vezes um o sis no meio do deserto Ainda que saiba que n o fui o aluno exemplar consegui realizar um sonho de longa data quando finalmente dos meus dedos tirei os Verdes Anos N o sei o que que no meu futuro haver de guitarra mas sei que me vou sempre lembrar das nossas aulas disso e de me ter que levantar s 6 da manh para a apanhar a porra do comboio para Haarlem e ainda ter que pedalar meia hora at sua casa Por fim julgo n o ter que me repetir para o Diabo visto que com excep o de Euskara ele fala todas as l nguas a menos que lhe chamemos Akerbeltz Bom e agora que aqui estou tomo a oportunidade para me dar a uns ares de rebeldia pois apercebo me que no fundo poderei escrever quase tudo o que bem me apete a neste ponto que a menos que algu m fa a uma tradu o tu sabes quem s quase ningu m vai perceber pelo menos n o as pessoas que poder o fazer algo sobre isso Posso ter posto a alma no penhor mas at data ela ainda minha e caso isso ainda n o tenha sido feito evidente ela a alma de um grandess ssimo punk Assim sendo passo de seguida a transcrever as proposi es desta tese que foram recusadas pelas altas inst ncias e que eram infinitamente melhores e mais inteligentes que as oficias 176 Entre outras coisas o m todo cientifico est des
60. age and know that there would be someone who would almost understand it Yibole if niceness was a disease you would be a terminal case Miao you re a good man and I can tell that deep down you re a fighter Maurits an advice no Japanese master will ever give you the perfect technique is not the strongest Van Thang I m sorry about the monster you re inheriting from me Giacomo you where only here for a short time but it was a most pleasant short time Jurgen Buschow your massively experienced insights are priceless Bert you were always available to tell some interesting vacation story whether we wanted to hear it or not Theo you came here as the archetypical Dutch student and I am happy I had the opportunity to watch you grow out of that socially imposed shell of mental mediocrity and into something of higher aspirations I wish you all the luck in Canada Of course I cannot in any way forget my three students Xinmin Peter and Mischa I have discussed student supervision with many of my PhD and Post doc colleagues and considering the stories they told me it seems like I completely lucked out with you three Your work was valuable beyond words and I am glad and proud to have known you and to have been able to put your names on three papers I wish you all rich and fruitful lives And even if what I write next may be considered inelegant I still feel like I have to write it Peter I honestly don t think this thesis would exist if it
61. agnetic anisotropy as the Mn Fe ssPooGeo sample c Magnetization versus applied magnetic field at 5 K for the Fe P azeu sample also demonstrating the same clear difference in magnetic behavior but giving a good insight on the influence of the small Mn content present on the Mn Fe Po9Geo sample The clear difference in magnetic behavior between the random powder and the field oriented powder suggests a strong anisotropy associated with an easy axis that allows for a rapid magnetic saturation Analyzing the results depicted in Figure 7 11 it can be concluded that indeed Ge is a fundamental element in the occurrence of this behavior Comparing the Mno Fe s Poo9Geo and the Mny Fe gsPogsGeo is samples we notice a reduction in magnetic anisotropy in the later while completely removing Ge appears to simply remove the occurrence of this behavior 29 giving room for other already studied 100 states of metamagnetism and antiferromagnetism related to the presence of small amounts of Mn in this system 27 30 The small amount of Mn in these samples does seem to play some part in this behavior given that the Fe o5Po9Geo sample while still possessing a clear magnetic behavior difference reaches its saturation magnetization much more rapidly in its random powder measurement than in the other samples This indicates that the small amount of Mn present in the other two samples does contribute to the large magneto crystalline anisotropy
62. agnetization and hard direction 1010 demonstrating a slow saturation for a close packed hexagonal Cobalt structure a b and c represent the basal plane and d an axis perpendicular to this plane in the abcd notation 3 13 2 2 1 Magnetocrystalline anisotropy There are several exchange interactions that play a part in magnetocrystalline anisotropy The first is the spin orbit coupling which causes the electronic orbitals to be influenced by electronic spin following it when magnetization changes orientation 2 Even though this is a clear source of anisotropy it is usually very weak being easily overcome by a few hundred oersteds 4 Shape anisotropy should also be mentioned as a relevant contributor although this is not developed in the current thesis This is the kind of anisotropy that arises from a magnetic body which does not have a perfect spherical shape This means that its demagnetization field will not be equal for all direction and in this case generate a non spherical energy surface of a shape complementary to the actual body with preferred directions Wplanes Even though this anisotropy phenomenon arises from purely geometrical contributions it cannot be disregarded or underestimated as it is the driving force behind such widely used magnetic materials as the Alnico Al Ni Co magnets which consist of FeCo rich needles in a NiAl rich matrix 2 5 The great contributor to anisotropy that will be focused upon is the
63. akes to fight for that happy ending Stepping outside of Delft I would like to mention Peter Foreshaw and Punita from Amsterdam University I know our meetings were brief and recent but our conversations gave me the strength to keep on pushing against the current I honestly hope we meet again in the proper and adequate circumstances Leaving the academies and going into the streets I would also like to thank Don Azito and the whole Amsterdam la Botanica crowd both living and dead It would be inelegant of me to start naming you particularly in this context but know that with your friendship openness and light you helped me more than you can ever imagine I managed to find my way out of some dark places thanks to you LUZ Also the tip of my hat and a gentleman s thank you to the Devil for having always stayed on that side of the mirror except that one time And finally to my Saints and Eguns even with all my sins for some reason I could never really understand someone in Heaven has always loved me Chega agora aquela parte em que se escreve algo na l ngua original do estudante em quest o E por favor note se que n o o fa o por muito apenas porque me parece que tal fica sempre elegante e misterioso aos olhos de quem a n o sabe ler quer dizer eu vi umas teses escritas por vietnamitas e aquilo impressiona muito s pena que o portugu s n o tenha mais assentos e sinais gramaticais Ainda assim j que aqui estou mai
64. al section in Springer Materials The Landolt B rnstein Database http www springermaterials com docs VSP datasheet Ipf c 00925000 LPFC_925042 html Y B Kuz ma E I Gladyshevskii Russ J Inorg Chem 9 1964 373 Co Mn Si ternary phase diagram isothermal section in Springer Materials The Landolt B rnstein Database http www springermaterials com docs VSP datasheet Ipf c 00925000 LPFC_925257 html T Krenke E Duman M Acet E F Wassermann X Moya L Mafiosa and A Planes Nat Mater 4 2005 450 T Krenke E Duman M Acet X Moya L Mafiosa and A Planes J Appl Phys 102 2007 102 P J von Ranke N A de Oliveira B P Alho E J R Plaza V S R de Sousa L Caron and M S Reis J Phys Condens Mat 21 2009 056004 29 30 31 32 R J Joenk J Appl Phys 34 1963 1097 A Biswas T Samanta S Banerjee and I Das J Phys Condens Mat 21 2009 506005 X Zhang B Zhang S Yu Z Liu W Xu G Liu J Chen Z Cao and G Wu Phys Rev B 76 2007 132403 S A Nikitin G Myalikgulyev A M Tishin M P Annaorazov K A Asatryan and A L Tyurin Phys Lett A 148 1990 363 85 86 Chapter 7 Magnetic and magnetocaloric exploration of Fe rich Mn Fe P Ge 7 1 Introduction As mentioned in Chapter 5 much of the MCE materials research performed in Delft focuses on promising Fe P type systems Among these is the Mn Fe gt P Ge system that ha
65. all measurement modules It is through this value that the data sets measured by each module are fitted together 4 Measurement frequency how often this equipment measures This box is present in every equipment block 5 Wellness indicators lighting either in green yellow or red Under normal circumstances these would light green otherwise they indicate an error These are present in every equipment block The Lakeshore 331 Temperature Controller block is displayed in Figure A 19 with the respective legend given in Table A 5 This block is crucial on all temperature dependent measurements and it is the one where one may more directly operate and control the setup during a measurement as all the others modules are fairly automated 129 Figure A 19 Lakeshore Temperature 331 Controller interface software block Table A 5 Legend for Figure A 19 1 On Off bottom 2 Read setpoint switch This switch determines if the measurement is reading and recording the setpoint temperature and the resistance of the carbon glass temperature dependent resistor from the Lakeshore 331 Temperature Controller module Sample space temperature in Kelvin This should be equally displayed on the front panel of the Lakeshore 331 Temperature Controller Setpoint temperature reading the temperature the Lakeshore 331 Temperature Controller is aiming at during heating Wcooling or a during a temperature ramp This sh
66. aloric refrigerator design has been suggested by Zhang et al 31 resorting to both regular magnetocaloric and inverse magnetocaloric materials in which cooling may be achieved by both the application and removal of a magnetic field in the refrigeration cycle making up a more efficient refrigerator Popular in this field of research are the RhFe 26 30 32 La Ca MnO 30 and Ni Mn Heulers alloys 26 27 In order to isolate the phase that produces these interesting results EDS measurements were performed on a set of samples In this way we established that the phase of interest corresponds to Mn Coo gSig gt Po s A Mn CoosSio Po s sample was produced and characterized in exactly the same terms as the ones belonging to the Mn Co Si P system X ray diffraction measurements revealed the sample was single phase and crystallizes in the orthorhombic Prima structure displayed in Figure 6 7 This structure is the same as the Co Fe P secondary phase found in the Fe Co Si P system Mn Fe Co atoms Si P atoms Figure 6 7 Puma orthorhombic structure of the Co Fe P system found as a secondary phase in the Fe Co Si P samples and the Mn CoosSio2Pos compound found in the Mn Co Si P system 81 This compound revealed a remarkably high magnetic transition compared with many compounds previously investigated for their inverse magnetocaloric effect 26 27 31 The magnetization reaching above 70 Am kg as shown i
67. and isolate the Fe P behavior in our data were taken from ref 21 For field values of 0 4 and 0 5 T there was a significant reduction in the sharpness and height of the transition peak indicating the change from a first order transition to a second order transition with Tc remaining constant Only for a field of 0 6 T did the transition temperature finally move up to 232 5 K Measurements performed with higher fields proved to be unreliable as the transition peak became smaller than our instrumental noise The behavior described above can be observed in Figure 8 3 109 0 4 Tesla Beer re 0 6 Tesla 0 7 Tesla 0 8 Tesla 176 184 192 200 208 216 224 232 240 248 Temperature K Figure 8 3 Fe P specific heat measurements taken in the experimental setup described in Chapter 4 with magnetic fields between 0 3 and 0 8 Tesla The specific heat values for solid copper used to calculate the contribution of our reference and isolate the Fe P behavior in our data were taken from ref 21 These results can be interpreted in the light of Caron ef al 10 but only if we consider that they show the results of a single crystal sample with an applied magnetic field along its hard direction perpendicular to the c axis This is a surprise as the sample used was supplied to us as being polycrystalline Further analysis of the results by Caron et al 10 does suggest that this might not be the case Figure 8 4 shows magnetization as a fun
68. arly for the 111 direction we will find Ra which B substituting in Equation 2 22 lead up the expression of E K ELS HK which then lead to K 2 4 E Enp 2 232 K Z E Eio 36 E Eio 2 23b e Torque Method This method requires the use of a torque magnetometer that can measure the torque N required to keep a crystal with its axis inclined at various known angles relatively to the external magnetic field A sample shaped like a disk or ellipsoid is rotated around an axis perpendicular to both its plane and the applied field which should be sufficient to saturate the sample In a cubic lattice the torque curves are expected to depend on the crystal plane of the sample A sample cut perpendicularly to the 001 direction will have values of a cos a sin and a 0 which when applied to Equation 2 22 results in Fan Ko K cos O sin 6 K 01 cos46 2 24 The differential of Equation 2 24 then gives us the torque as NE findo 2 25 2 20 It should be noted that Equation 2 24 is specific for the cubic case Other expressions will need to be derived from the anisotropy energy equations corresponding to other structures Having the above specified torque measurements as exemplified by Figure 2 6 after a Fourier analysis of the corresponding curves and comparison to Equation 2 25 or any other obtained from of the structure s anisotropy energy expression one may arrive at the
69. asurements to calculate AS Equation 2 5 The calculation of AT on the other hand is slightly more complex as it relies on isofield specific heat measurements as well as the same magnetization measurements required for the calculation AS equation 2 10 1 Even though for a proper study of the magnetocaloric potential both these quantities are required 2 3 the calculation of AT is usually made difficult by the lack of a commercial measurement systems that allows reliable specific heat measurements in applied magnetic field To overcome this issue many research groups world wide resort to assembling their own isofield specific heat measuring equipments Among the most recent examples we may cite the setup described by Marcos et al 4 consisting of an insert that can be fitted to any cryostat with the capacity to generate a magnetic field This setup resorts to thermo batteries which give a voltage output in response to the heat exchange with the measured sample One other example described by Korolev eft al 5 has been specifically designed to measure magnetic colloids Instead of using a permanent magnet it is designed as a microcalorimetry cell placed between the two poles of an electromagnet to generate a low intensity magnetic field The setup described by Kuepferling et al 6 on the other hand resorts to commercial Peltier cells a thermoelectric device made of a series of junctions of conductors with different thermoelectric
70. ation std ratio relative std deviation std deviation average heat capacity A and B mean valA mean valB Sprint results heat capacity A and B stat dev std valA std valB Sprint deviations 139 ratio std relstd average std deviation relative std deviation print ratio capacity temperature matcher m Parameters amount_to_average 10 filename temperature csv first column time second column thermopile voltage measurements CSV Seperator the seperator in the csv files Define data variables data dlmread filename csv seperator t time data 1 t value data 2 t index zeros size ratio time for index 1 1length ratio time time ratio time index t index index find t time gt time 1 first end ratio temperature zeros size t index for index 1 1length t index 1 t index index ratio temperature index mean t value i amount to average 2 itamount to average 2 end A 3 4 Data display The software used for this step in particular is not specific to the microcalorimetry instrument and data and as such will not be discussed in detail at this point A regular software like Origin is more than enough to handle this step 140 A 4 Measurement preparation protocol What follows are step by step instruction for the handling and mounting of both measurement and references samp
71. ayed in Figure 7 4 This however turned out to be challenging as in this concentration range properties such as sharpness or magnetic behavior are extremely sensitive to small compositional changes In this sense while maintaining Ge 0 12 we have determined the Fe maximum and consequently the maximum Tc in which this system still displays usable characteristics for a magnetic cooling device as shown in Figure 7 8a 95 Mn Fe P Ge m xz1 43 4 xz1 45 v x 1 47 x 1 49 4 x 1 51 1 95 x x 0 88 1 Tesla Magnetization Am kg 100 120 140 160 180 200 220 240 260 280 300 Temperature K AS JikgK 400 120 140 160 180 200 220 240 260 280 300 Temperature K Figure 7 8 a Magnetization versus temperature curves for Mn Fe Po ssGeo 12 samples for an applied magnetic field of 1 Tesla demonstrating the change in Tc with increasing Fe content b Magnetic entropy change for Mn Fe P ssGeo 12 samples for an applied magnetic field of 2 Tesla We have thus determined a maximum T of about 225 K for a maximum Fe content of 1 51 for the determined Ge minimum 0 12 Due to the partial loss of magnetic transition sharpness when compared with Mn rich compounds the magnetic entropy change calculated for these samples displayed in Figure 7 8b is not as high as those measured previously by Trung et al 1 but are comparable with those observed for pure Gd metal
72. ces of 99 5 purity As the current samples no longer contain elemental phosphorous an arc melting furnace was used for their preparation The resulting metallic buttons were annealed in the exact same conditions as the samples belonging to the Mn Fe Si P system The characterization of the samples was conducted in the exact same way as those described in section 5 3 2 5 6 2 Results and discussion X ray diffraction measurements on boron doped samples with boron concentration of 0 05 0 1 and 0 15 revealed that all of these contain a minor Si peak in contrast to the undoped samples The intensity of this peak remained stable for these three concentrations suggesting a partial substitution of S1 by boron in the sample s crystal lattice as opposed to a purely interstitial fit as demonstrated by Figure 5 8 63 n S MnFe SiB MnFe Si 4000 Si 3000 2 5 o 2000 O 1000 0 30 35 40 45 50 55 60 20 degrees Figure 5 8 X ray diffraction pattern for the sample MnFe SiB revealing the cubic Fm3m structure of the Si rich Mn Fe Si P system and minor peaks consistent with pure Si segregation Samples with a boron concentration higher then 0 15 were found to exhibit besides the Si segregation other minor multiple secondary phases The exact nature of these phases was not investigated in depth as a preliminary observation of the X ray diffraction patterns of these samples revealed that the formation of these s
73. compound FejP but can easily be tuned towards more practical ends given its particular sensitivity to stoichiometric changes and synthesis methods The magnetic properties of Fe P itself have been widely studied in the past for permanent magnet applications as mentioned in Chapter 7 or for pure academic interest Still a great deal of uncertainty and speculation has hovered over this compound precisely given its extreme sensitivity to stoichiometry and to external magnetic fields which give rise to incompatible and disparate results in different publication 8 9 Given the importance currently placed on this compound there is a need for a complete assessment and understanding of the specific heat of Fe P under a magnetic field These measurements are inspired by the recent work performed by Caron ef al 10 8 2 Material overview Fe P crystallizes in the already discussed hexagonal P 6 2m structure We have two magnetic sub lattices both occupied by Fe Fe I in the tetragonal 3f and Fe II in the pyramidal 3g site and P the dissimilar 2c and b sites atoms 10 11 In this arrangement Fe I and Fe II have a total moment around 2 9 up f u with Fe I having a moment close to 1 up f u and Fe II close to 2 up f u 10 12 14 Information regarding moment orientation can be found in Chapter 7 section 5 Besides its first order ferro to paramagnetic transition at 217 K this compound also exhibits a metamagnetic transition at relatively low
74. compounds by resorting to sulfur substitution given the periodic table proximity of this element with elemental phosphorus All samples used for this study were produced and characterized in the same way as all others belonging to the Fe Co 3 Si P system The used S was 99 5 pure powder from Alfa Aesar X ray diffraction patterns revealed that all S containing samples show the tetragonal 14 structure without the occurrence of any second phase The same S substitution causes a decrease in both lattice parameters a and c with a 9 0263 and c 4 4504 A for Fe Co 4P and a 9 0192 A and c 4 4499 for Fe Co jPosSo All S containing samples show a high magnetization value of 110 Am kg Similarly to other Fe and P rich samples Tc was still beyond the reach of our magnetometers but DSC measurements indeed demonstrate a slight decrease in the transition temperature DSC measurements further display an unexpected second unidentified transition around 372 K as shown in Figure 6 4 In this new Fe Co s P S system the critical temperature similarly to the Fe Co P system decreases with increasing Co content 1 1 n gt Es N oo to o Normalized Specific Heat e o u 300 375 450 525 600 675 Temperature K Figure 6 4 Comparison between the normalized specific heat of the Fe Co P and the Fe1 6C01 4P0 950 samples demonstrating the slight decrease in Tc with S substitution and the occurrence of a new and unidentif
75. control the temperature of the sample space while the chip heaters merely provide a heat pulse to the sample reference which enables for the performance of relaxation measurements All the wires described above except those referring to the samples space heater s h1 and B connect the chip sockets and resistor to the 16 pin insert head plug The wire numeration used in the above Figures and Tables relates to this point as displayed in Figure A 13 2 4 6 8 1012 14 16 rdl EMMA M MMM MM Pg 1 3 Sy Sa 28235 Figure A 13 Wire layout on the insert head plug Not all pins in this plug are used in the current setup 124 A 2 2 Insert body and top Rising from the insert head the sample space wires are bundled together with the sample space heater wires and travel along the body of the insert to two cable plugs D1 and D2 on the insert top Each of these plugs has 16 pins meaning that the wires coming from the insert head and heaters are distributed among both of them being that some wires are actually repeated on both plugs The layout of these plugs is displayed in Figure A 5 Figure A 14 Wire layout on the plugs at the top of the insert A 2 3 Insert cables Connecting the D1 and D2 plugs to the several measurement modules on the outside of the setup are two cables equality designated as D1 and D2 according to the their corresponding plug The wire layout in these as is logical is a mirror image of Figure A 14
76. ction of the magnetic field on polycrystalline and single crystalline Fe P both parallel and perpendicular to its easy direction 110 100 3 80 j x E Ss 60 li c ig 40 j cLugH polycrystal 20 T 10K al 0 1 2 3 4 5 HoH T Figure 8 4 Magnetization vs magnetic field measurements on polycrystalline and single crystal Fe2P both parallel and perpendicular to its easy direction 10 It should be noted that the results given by polycrystalline Fe P mimic the ones by the single crystal measured with a field parallel to the easy axis which should the sample in question be truly polycrystalline is not expected What seems to be the case is that all samples taken to be polycrystalline are in fact single crystals or near single crystals which given their small size have a clearly predominant easy direction Should our case be that of a truly polycrystalline sample considering that this would have its various grains crystals alighted along every possible direction one would expect an initial magnetization value of 5096 of that of a pure single crystal The values for the polycrystalline sample in Figure 8 4 on the other hand amount to nearly 8396 of the magnetization of the single crystal signifying that the various grains crystals composing this sample have an average misalignment of 34 degrees with the applied magnetic field Should this hypothesis be correct it sheds a great deal of light on our current
77. d be regarded as completely separate and independent As such the application of a magnetic field on the sample space while this is inside the cryostat needs to be done manually and externally after which one may proceed to with a measurement as described above Given that the application of a magnetic field by the Model 430 Power Supply Programmer is not specific to the current microcalorimetry instrument you should refer to this module s User s Manual for more details 154 Currently this instrument is not fitted for the registration of magnetic field sweeps A 5 4 Stopping a measurement Depending on you measurement there may be various degrees of complexity to this step in particular Approaching the interface software press all the On buttons in exactly the opposite order as indicated in Figure A 26 with the exception of the Labview on off button which should be substituted by the large STOP button on the interface software saving block as indicated in Figure A 32 Figure A 32 Location of the STOP button on the Saving block of the interface software The above process may take a few minutes to conclude depending on the length of the measurement Besides this you should also turn off the heater output of the Lakeshore 331 Temperature Controller as shutting down the interface software will not do this automatically Press Remote Local followed by the Heater Off button on the Lakeshore 331 Temperature Controller to do
78. d zekere fundamentele eigenschappen van deze systemen te bestuderen en eveneens een economische optimalizatie van het Mn Fe 3 P Ge systeem Na een algemene en contextualiserende introductie in hoofdstuk 1 vervolgt het in hoofdstuk 2 met de behandeling van theoretische aspecten van het magnetocalorisch effect evenals de beschrijving van het magnetisch anisotropie fenomeen waarbij beide concepten fundamenteel zijn voor begrip van de resultaten gepresenteerd in de volgende hoofdstukken Al de experimentele procedures voor monster bereiding en karakterisatie worden gedetailleerd gegeven in hoofdstuk 3 met de details van de experimentele microcalorimetrie in hoofdstuk 4 Dit soortelijke warmte meetinstrument met de mogelijkheid van het toepassen van een magnetisch veld tot 9 Tesla hang af van micro calorimetrie chips van het bedrijf Xensor Integration Met dit instrument is het mogelijk soortelijke warmte metingen te doen in een magnetisch veld tot 9 Tesla aan milligram monsters Dit biedt de mogelijkheid de ware adiabatische temperatuur verandering van een materiaal te berekenen evenals betrouwbare en precieze informatie te geven over enige fase overgang die be nvloed wordt door een magnetisch veld In hoofdstuk 5 het Mn Fe Si P systeem is volledig onderzocht met R ntgen Diffractie DSC en magnetisatie metingen Van dit systeem is vastgesteld dat het niet geschikt is voor magnetocalorische toepassingen maar het gedane onderzoek liet toe e
79. e Keithley 2000 4 Measurement frequency how often this equipment measures This box is present in every equipment block but it is of an essential importance on this block as the calculation of the relaxation measurements performed requires a large number of data point 5 Wellness indicators lighting either in green yellow or red Under normal circumstances these should light green otherwise they indicate an error These are present in every equipment block 6 Pause indicator Should any number be introduced in this box the software will pause the whole system for that amount of time whenever it changes channel This was built so as to prevent the measuring of aberrant first measurements caused by channel switching Given that the calculations to follow resort merely to the temperate relaxation curve this number is not important The Keithley 2002 block is displayed in detail in Figure A 18 with the respective legend given in Table A 4 This module doesn t play a significant role in the current measurement procedure 128 Figure A 18 Keithley 2002 interface software block Table A 4 Legend for Figure A 18 On Off button 2 Difference between the thermopile voltages of the two Xensor microcalorimetry chips This is equally displayed on the front panel of the Keithley 2002 3 Time indicator This value is measured by an external clock and as such is the same for
80. e often disagreed on many topics but you always showed yourself patient and graceful with my shortcomings more than I probably would if I was ever in your position We once again fall into a clich here I will always be thankful for having been given the opportunity to study in this strange and faraway land I have come to love And this is not because of the academic or scientific work I developed here but rather becouse coming to the Netherlands enabled me to achieve more of myself than I would care to divulge Niels should also be mentioned at this point Although you were technically my co supervisor this was actually something I only discovered during my second year I think we very rarely had any work related conversations Even still you are a most pleasant and intelligent man and all our intercalations be them work related or not were always exceedingly interesting and entertaining Of course that now following the proper etiquette I should mention my colleagues of the Delft magnetocalorics research group From the very beginning you were all welcoming and certainly aided with my learning curve Ou I think we shared too many hotel rooms Trung your good mood and hard work were always an inspiration Anton know that you are my favorite Dutchman Frangois it felt good to finally not be the only European student around here Luana doesn t count Luana it is hard to express how much of a luxury it was to be able to speak my own langu
81. ease in magnetization with increasing Fe content abrupt jumps in magnetization were clearly observed in Figure 5 6 This result can be partially explained by site preference of the Mn and Fe atoms described in section 5 2 1 56 m x 2 9 x 2 3 A xz2 1 lt x 2 4 v x72 2 x 2 5 0 01 Tesla m x 2 Magnetization Am kg L Py Magnetization Am2 kg 2o N 50 200 250 300 350 400 Temperature K 0 50 100 150 200 250 300 350 Temperature K Figure 5 6 Magnetization versus temperature plot for Mns Fe Si with x between 2 and 2 5 revealing the possible effects of the site preference of Mn and Fe atoms described in section 2 1 in magnetization Insert magnetization versus temperature measurements for low magnetic fields 0 01 T revealing the change in Tc with Fe content 5 5 The Mn Fe Si P magnetostructural phase diagram By compiling all the data from our X ray diffraction patterns magnetization measurements and DSC measurements together with literature data from 12 14 19 20 and 21 we can construct the magnetostructural phase diagram of the Mn Fe 3 Si P system shown in Figure 5 7 Details regarding the samples can be consulted in Table 5 1 57 8S 0 1 0 2 03 04 05 06 07 08 09 10 TT K 22 00 73 13 124 3 175 4 226 5 277 6 328 8 379 9 431 0 482 1 533 3 584 4 635 5 686 6 737 8 788 9 840 0 Current Wo
82. econdary phases did not appear to follow a particular trend or logical evolution with increasing boron content as demonstrated in Figure 5 9 The lattice parameter a of our doped samples increased for a boron content of 0 02 while for higher concentrations it was found to decreases in a non linearly fashion These results further underline that indeed B atoms do not enter this system as an interstitial element above the content of 0 02 but are partially substituting Si in the crystal structure which effectively limits the amount of boron that can be added to this Structure 64 2000 Mn Fe SiB 1750 Mn Fe SiB Mn Fe SiB 4 5 1500 1250 1000 Counts 750 500 250 Position 20 Copper Figure 5 9 Comparison of the X ray diffraction patterns of the samples Mno Fe SiBo3 MnooFe SiBos and Mn Fe SiB s revealing an apparent random occurrence of secondary phases as a result of increasing B content among the cubic main phase 5 6 3 Magnetic results The antiferromagnetic ferromagnetic transition temperature was found to increase with boron concentrations of 0 02 and 0 05 With the added boron these samples also presented a slightly higher magnetization and a sharper ferromagnetic paramagnetic transition For higher boron concentration however the transition temperature not only decreased but we also observed a loss of sharpness in the antiferromagnetic ferromagnetic transition and an increase
83. eis que um dem nio realmente se queira dar ao trabalho de falar com elas 177 178 List of Publication Papers Published before the current PhD 1 J V Leit o D L Rocco J S Amaral M S Reis V S Amaral R P Fernandes N V Martins and P B Tavares Influence of the Magnetic Anisotropy on the Magnetic Entropy Change of Ni Mn Ga Bi Memory Shape Alloy IEEE Transactions on Magnetics 44 2008 3036 2 D L Rocco J S Amaral J V Leit o V S Amaral M S Reis R P Fernandes A M Pereira J P Ara jo Nuno V Martins P B Tavares and A A Coelho Percolation processes and spin reorientation of PrNis Coy Physical Review B 79 2008 014428 3 E J R Plaza V S R de Sousa P J von Ranke A M Gomes D L Rocco J V Leit o and M S Reis A comparative study of the magnetocaloric effect in RNiz R Nd Gd Tb intermetallic compounds Journal of Applied Physics 105 2009 013903 4 D L Rocco J S Amaral J V Leit o V S Amaral M S Reis Soma Das R P Fernandes J P Ara jo A M Pereira P B Tavares Nuno V Martins and A A Coelho High refrigerant capacity of PrNis Co magnetic compounds exploiting its spin reorientation and magnetic transition over a wide temperature zone Journal of Physics D Applied Physics 42 2009 055002 Published during the current PhD 1 J V Leit o You Xinmin L Caron and E Br ck Magnetostructural study of the Mn
84. en magnetostructurele kaart van het systeem samen te stellen Naast de reeds bekende cubische fase voor Mn Fe Si systeem en de tetragonale orthorhombische fasen van het Mn Fe P systeem een nieuwe hexagonale fase voor Mn Fe Si P werd gevonden in het gebied met ongeveer 0 2 lt x lt 2 0 en 0 2 lt y lt 0 9 Magnetisatie metingen bevestigen en detailleren de al bekende eigenschappen van de Mn Fe Si en Mn Fe P systemen Gegeven het interessante karaker van de antiferro ferromagnetische overgang zoals aanwezig in Si rijke monsters met een Fe gehalte van 1 2 tot 2 3 een studie van het effect van interstitieel borium werd ook gedaan op dit systeem Dit resulteerde in het verschuiven van de overgang van 50 5 naar 57 K bij een B gehalten van 0 05 terwijl bij alle hogere concentraties secondaire fases ontstonden Hoofdstuk 6 is een poging de nieuw ontdekte hexagonale fase in het Mn Fe 3 Si P systeem te bestuderen door de introductie van Co het eerder bestudeerde systeem en op deze manier Mn Co s Si P en Fe Co 3 Si P te cre ren Resultaten waren voor geen van deze systemen overtuigend aangezien ze geen van beide als enkele fase lijken te bestaan Nochtans de exploratie van verscheidene secondaire fases zoals gevonden in de Mn Co Si P monsters onthulden het bestaan van het Mn Co o Si P systeem dat een extreem relevant invers magnetocalorisch effect liet zien Een poging Si vrij Tc af te stemmen in het Fe Co P systeem lie
85. enhado para remover o individuo circunstancial do processo de investiga o cientifica no entanto not vel ver a quantidades de egos que florescem neste meio O paradigma p s moderno do investigador profissional procura produzir um individuo ignorante amoral e desumanizado Tal fera um perigo iminente para toda a humanidade e dever ser abatida primeira vista relev ncia da pol tica inter institucional amea a eclipsar real m rito cient fico Ci ncia Russa n o ci ncia normal Dado o tempo que os seres humanos existem neste mundo como tal estranho que a morte seja ainda encarada como uma surpresa e uma trag dia Bergson pode ser considerado brilhantes mas o seu Mati re et m moire uma obra ignorante e mal informada uma grosseira m interpreta o e subestima o do que a consci ncia e apenas poderia ter sido escrita por um Franc s O Diabo o reflexo de um homem noite Em pr ticas Goeticas os dois paradigmas opostos de mera proje o mental subjetiva popularizada pela introdu o de Crowley ao The Book of the Goetia of Solomon the King e a exist ncia objetiva de uma entidade exterior ao carcista largamente proclamada pelo corrente Revivalismo M gico s o ambos conceitos egoc ntricos A maioria das mentes n o tem capacidade para projetar uma alucina o coerente que possa ser interpretada como um dem nio e a maioria das pessoas n o s o assim t o not v
86. ent of 1 9 to 2 35 the cubic phase borders the orthorhombic phase As we further increase the Fe content Tc of the cubic phase also continues to increase until it reaches its highest value of 840 K in the Fe3Si compound 59 On the P rich side the orthorhombic phase reaches its border at an Fe content of about 2 1 and the tetragonal 14 structure is once again observed until the complete substitution of Mn by Fe In this area the tetragonal phase now displays a transition between a ferromagnetic and a paramagnetic state with an Fe dependent increase in Tc until it reaches its maximum of 700 K in the Fe P compound In the Fe rich region the cubic and tetragonal phases display a border along the Mns Fe Sio 4Po with 2 35 x 3 All paramagnetic ferromagnetic phase transitions observed in this system present the characteristics of second order phase transitions Table 5 1 Details regarding crystal structure magnetic behavior and transition temperatures for every sample produced for the current chapter Sample Composition MnsSi Mn sFeo sSi Mn FeSi Mn sFe 5Si MnFe Si Mno abe Si MnoFe Si Mna be 381 MnoFe 4Si Mno sFe sSi Fe3Si Mn3Sio sPo 2 Mn Feo 5Sio sPo 2 Mn FeSio gPo2 Mn sFe sSio sPo 2 MnFe2Sig gPo 2 Mna she sSio sPo 2 FesSio sPo2 Mn Fe 8Sio cPo 4 Mn Fe oSiocPo4 60 Structure at Room Temperature Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m Fm3m I
87. erbinding zien bij een toenemend toegepast magnetisch veld welke bij een kritische waarde van 0 4T 170 arriveert als de overgang geheel 2e orde is Een verdere temperatuur verschuiving van deze overgang van ongeveer 1 5 K wordt waargenomen bij een veld van 0 6 T Deze resultaten waren eigenlijk nogal onverwacht en worden alleen duidelijk als we overwegen dat het gemeten monster vrijwel een enkel kristal is gepositioneerd in de opstelling langs de harde as Meer metingen zijn nodig op deze zaak op te helderen Tenslotte is de Addendum een gedetailleerde stap voor stap bedienings handleiding voor de experimentele installatie zoals beschreven in hoofdstuk 4 en gebruikt in hoofdstuk 8 Deze handleiding beschrijft in details alle relevante aspecten van de opstelling van hardware tot software en dekt het hanteren de bedrading monster behandeling en meting en gegevens verwerking 171 172 Acknowledgements As it is somewhat of a clich I can never expect to thank all the people who have helped me along these four years of PhD work Furthermore all those who are relevant and close to my heart know who they are and they shouldn t have the necessity to have their names publicized on a book practically no one will ever read None the less I believe traditions exist for a reason and I ll make an effort to follow yet one more As is polite I should naturally start with my supervisor Prof Ekkes Br ck Whether these were expressed or not w
88. erved in the cubic Si rich samples with Fe content ranging from 1 2 to 2 3 To this end a series of samples where produced containing small amounts of B with the intention of having this element enter interstitially into the cubic lattice and in this way possibly move the antiferromagentic transition up from its value of Ty 50 K Such a method had been previously used with great success in magnetocaloric studies for the control to key properties like Tc magnetization and magnetic entropy change in La Fe Si systems 23 shape memory alloys 24 and MnCoGe systems 25 In contrast to the usual stoichiometric tuning the technique of inserting interstitial atoms in a crystal lattice is meant to solely affect the inter atomic distance and not the electronic interactions between atoms directly via the valence electron concentration e a 24 62 For this end a series of Mn Fe SiB samples with x 1 9 2 2 1 and 2 2 and y 0 02 0 05 0 1 0 15 0 2 0 3 0 4 and 0 5 were produced to study the tuning potential of boron doping on the antiferromagnetic ferromagnetic phase transition of the cubic Mn Fe 3 Si P phase 5 6 1 Sample preparation and characterization All samples produced for the study of the effects of interstitial boron were made from the same starting materials as those used in the preparation of samples belonging to the Mn Fe Si P system and previously described in section 5 3 1 The boron used consisted of crystalline pie
89. es not belong to the Fe Co 3 Si P system but most likely to the Co Fe P system To confirm this hypothesis we produced a sample belonging to this Co Fe P system and found its X ray diffraction pattern to be consistent with the second phase present in our P samples as can be observed in Figure 6 1b 73 FeCo Si 2 z 2 o o Cubic Fm3m Orthorhombic Pmna 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Position 20 Copper b 2100 FeCo P 1750 S Cubic Fm3m 1400 Orthorhombic Pmna 1050 700 E 350 AA M ME EP RN TRI cle pital E 3750 o o 20 25 30 35 40 Position 20 Copper Figure 6 1 a Comparison of the X ray diffraction pattern of the FeCo Si compound demonstrating a pure cubic single phase and the FeCo Si P compound demonstrating the rise of a secondary phase consistent with the orthorhombic Pnma structure b Comparison of the X ray diffraction pattern of the FeCo P compound demonstrating both the Cubic Fm3m and the orthorhombic Puma structures and the FeCoP compound demonstrating a pure single phased Pmna structure which corresponds perfectly to the secondary orthorhombic phase observed in the FeCo P compound In terms of magnetic behavior the FeCo Si sample was found to have a high saturation magnetization of about 140 Am kg With increasing P content the saturation magnetization steadily decreases due to the formation of the alread
90. eterodox Exposition Mandragora Further Explorations in Esoteric Poesis Scarlet Imprint Croydon UK 2012 3 Jos Leit o Clavis Cyprianus Nova guia 10 2012 92 4 Jos Leit o The Grimoire of Saint Cyprian or The Prodigies of the Devil Conjure Codex Hadean Press Vol 1 n 2 2013 78 5 Jos Leit o Deambula o Nova guia 11 2013 47 6 Jos Leit o The folk and oral roots of the Portuguese Livro de S o Cipriano Sharing Cultures 2013 Proceedings 2013 265 7 Jos Leit o The folk and oral roots of the Portuguese Livro de S o Cipriano International Journal of Heritage and Sustainable Development invited publication based on the proceedings of the Sharing Cultures 2013 conference Submited 8 Jos Leit o translation and comments The Book of Saint Cyprian The Sorcerer s Treasure Hadean Press Accepted for publication in Spring 2014 and currently in editing 9 Jos Leit o Nauts and Nymphs Clavis Journal of Occult Arts Letters and Experience submitted 10 Jos Leit o Unapologetically Catholic Use and Practicality of the Cult of the Saints Under preparation 181 11 Jos Leit o translation and comments Bibliotheca Valenciana The Collected Works of Jeronimo Cortez Under preparation 182 Corriculum Vitse Jos Carlos Vieira Leit o was born in 1984 in Leiria Portugal In 2002 he entered into the Universit
91. etween atoms 91 Mn Fe PP Ge Mn Fe P Ge Mn Fe P Ge 1 2 0 85 1 2 0 8 0 15 0 16 0 17 0 18 0 19 0 20 31 20 Tans Temp K 220 0 240 0 260 0 280 0 300 0 320 0 340 0 360 0 380 0 400 0 420 0 440 0 d 1 25 d 1 30 d 1 35 1 40 5 d 1 40 Fe content x 1 45 5 d 1 45 1 50 A 1 50 0 15 0 16 0 17 0 18 0 19 0 20 Mn Fe P Ge 45 Ge content y Mn Fe P Ge Figure 7 4 Evolution of Tc with both Fe and Ge content in the Mn Fe P Ge system The open dots mark the compositions of the samples used in this study The bar on the right of the figure represents the tansition temperature of the composition range As can be seen in Figure 7 4 our initial hypothesis to reduce Ge by increasing Fe content while maintaining Tc around room temperature was correct This process has however also brought on an undesirable widening of the typically sharp transition between the Ferromagnetic and the Paramagnetic states in this system which no longer displays the characteristics of a first order phase transition Such an occurrence is extremely unfortunate for the prospect of applying Fe rich Mn Fe P Ge samples to any practical magnetic cooling device Figure 7 5 illustrates the above mentioned results 92 gt Mn re P Ge 1 25 0 8 4 Mn re P E 1 25 0 85 0 15 1 Tesla 140 120 100 80
92. f Equations 2 3a and 2 3b we obtain 2 4 Ho oH oT H l The change in entropy AS from an initial magnetic field H to a final magnetic field of H thus corresponds to ASD a4 uj H H amm o I dH 2 5 this results in the expression for the calculation of magnetic AS From this expression we can conclude that the calculation of the magnetic entropy AS of a given magnetic material can be achieved thought the measurement of magnetic isotherms It can be easily observed that this quantity will be maximized around large variations in magnetization with temperature as those that happen around the Curie temperature Tc the critical temperature at which a ferromagnet turns into a paramagnet It is obvious that in the search for novel magnetic materials for commercial cooling applications one should focus on materials with sharp transitions in the vicinity of room temperature the temperature we wish to cool from Following from Equation 2 3a through the first law of thermodynamics we can calculate the specific heat of a system with the second derivative of the Gibbs free energy a E Gem mis 2 6 d E en 01 Jens Considering entropy as a function of temperature and magnetic field S S T H a small change is represented as os os dS dT dH 2 7 5 5 2 10 Considering an adiabatic process dS 0 we obtain AS as 28 ar 8 an Taking Equation 2 6 2 4 and
93. g Dutput a ures Rag ocr Source Level Delay2400 m5 1 Sh e Bino food Current Source Level Sweep Startvalse Dday StopValue Points Figure A 20 Keithley 2400 interface software block 131 Table A 6 Legend for Figure A 20 On Off button of the square wave generator 2 OnVOff button of the reading functionality of the source meter can only be turned on when either buttons 1 or 3 are turned on 3 On Off button of the sweep function In the current setup there is no foreseeable use for this feature 4 Xensor microcalorimetry chip resistance and voltage registered in response to the wave generated by this module 5 Intensity registered in response to the square wave generated This should be equally displayed on the front panel of the Keithley 2400 Source level indicator See next point 7 Source level and source range setters These refer to the characteristics of the square wave being generated by the Keithley 2400 currently an 1 V square wave Should these values need to be changed this can only be done with some knowledge of Labview 8 Measurement frequency This it is present in every equipment block Wellness indicators lighting either in green yellow or red Under normal circumstances these would light green otherwise they indicate an error These are present in every equipment block The saving block is displayed in Figure A 21 with
94. ga H Fujii and T Okamoto Jpn J Appl Phys 16 1977 2175 83 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 84 Hui ping Liu P James A Broddefalk Y Andersson P Granberg and O Eriksson J Magn Magn Mater 189 1998 69 R J Gambino T R McGuire and Y Nakamura J Appl Phys 38 1967 1253 R Grasin E Vinteler A Bezergheanu C Rusu R Pacurariu I G Deac and R Tetean Acta Phys Pol A 118 2010 648 B H lsen M Scheffler and P Kratzer Phys Rev B 79 2009 094407 G G koglu and O G lseren Eur Phys J B 76 2010 321 S Picozzi A Continenza and A J Freeman Phys Rev B 66 2002 094421 S Ishida T Masaki S Fujii and S Asano Physica B 245 1998 1 B Balke G H Fecher H C Kandpal C Felser K Kobayashi E Ikenaga Jung Jin Kim and S Ueda Phys Rev B 74 2006 104405 S V Oryshchyn T I Berezyuk Y B Kuz ma Russ J Inorg Chem 32 1987 295 Z Yao S Gong J Fu Yue Sheng Zhang and Kai Lun Yao Solid State Commun 150 2010 2239 Fruchart A Roger and J P Senateur J Appl Phys 40 1969 1250 S Kumar S Chander A Krishnamurthy B K Srivastava J Magn Magn Mater 237 2001 135 S Kumar A Krishnamurthy and B K Srivastava J Phys D Appl Phys 41 2009 055001 V Raghavan Indian Inst Met 3 1988 51 Co Fe P ternary phase diagram isotherm
95. h of the material research being done in Delft on magnetocaloric materials a series of microcalorimetry measurements were performed in a polycrystalline Fe P sample in the experimental setup described in Chapter 4 112 Our results were unexpected and could only be understood if we consider that the sample measured was a single grain positioned in the equipment along its hard axis References 1 2 3 4 5 7 9 10 11 12 13 14 15 N T Trung Z Q Ou T J Gortenmulder O Tegus K H J Buschow and E Br ck Appl Phys Lett 94 102513 2009 N T Trung J C P Klaasse O Tegus D T Cam Thanh K H J Buschow and E Br ck J Phys D Appl Phys 43 2010 015002 N H Dung L Zhang Z Q Ou and E Br ck Appl Phys Lett 99 2011 092511 N H Dung Z Q Ou L Caron L Zhang D T Cam Thanh G A de Wijs R A de Groot K H J Buschow and E Br ck Adv Energy Mater 1 2011 1215 E Br ck N T Trung Z Q Ou and K H J Buschow Scripta Mater 67 2012 590 N H Dung L Zhang Z Q Ou L Zhao L van Eijck A M Mulders M Avdeev E Suard N H van Dijk and E Br ck Phys Rev B 86 2012 045134 N H Dung L Zhang Z Q Ou and E Br ck Scripta Mater 67 2012 975 H Fujii T Hokabe T Kamigaichi and T Okamoto J Phys Soc Jpn 43 1977 41 L Lundgren G Tarmohamed O Beckman B Carlsson and S Rundqvist Phys Scri
96. have been studied in the past such as spintronics research in its ferromagnetic phase or on metal metalloid compound research an area of interest in nuclear reactor materials science 67 Further experimentation with interstitial boron was performed in an attempt to tune the antiferromagnetic ferromagnetic transition observed in the Si rich cubic phase of the Mn Fe Si P system The antiferromagnetic transition could be enhanced by addition of low boron concentrations but it decreases for higher additions This system was found inappropriate for magnetocaloric applications This research and the understanding it offers may however open the possibility for further magnetocaloric studies and developments The novel hexagonal phase still represents a new and unexplored set of compounds that may be optimized by the addition of a fifth element References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 68 N H Dung L Zhang Z Q Ou and E Br ck Appl Phys Lett 99 2011 092511 N H Dung L Zhang Z Q Ou L Zhao L van Eijck A M Mulders M Avdeev E Suard N H van Dijk and Ekkes Br ck Phys Rev B 86 2012 045134 S Tomiyoshi Y Yamaguchi M Ohashi E R Cowley and G Shirane Phys Rev B 36 1987 2118 L Hongzhi Z Zhiyong M Li X Shifeng L Heyan Q Jingping L Yangxian and W Guangheng J Phys D Appl Phys 40 2007 7121 S T
97. hematically shown in Figure 3 5a The great sensitivity of SQUID devices originate from the detection of changes in magnetic field associated with one flux quantum 5 2 067833636x 10 Wb 3 1 29 With a constant biasing current in the SQUID device the measured voltage will oscillate with the changes in phase at the two junctions which depends upon the change in the magnetic flux Counting the oscillations allows for the evaluation of the flux change Magnetic measurements were thus performed in two different SQUID magnetometers a MPMS 5S and a MPMS XL model both from Quantum Design The measurements taken were unless mentioned otherwise temperature sweeps from 5 to 370 K MPMS 55 or 400 K MPMS XL with a fixed applied magnetic field SAMPLE CHAMBER TUBE SAMPLE SPACE 150 TEERMAL SHEET WITH INNER VACUUM JACKET WALL N HEATER ANNULAR COOLING REGION swa OUTER VACUUM JACKET WALL SECOND DERIVATIVE COIL Magnetic field SUPERCONDUCTING WIRE Superconductor Biasing Biasing current current SAMPLE Josephson MILTIFLAMENT PICK OP CONS SUPERCONDUCTING WIRE on composite lara One period of 4 voltage variation lt 2 corresponds to 7 H an increase of one flux quantum COMPOSITE FORM FOR SOLENOID ion leasing Varias ily 5 x tic f ge Cadi al MAGNETIC FIELD H Volta Or St Magn L Figure 3 5 a Schematic representation of a Josephs
98. hich the Keithley 2000 is measuring a string of values of either 3 or 4 2000 volt Voltage response from the Keithley 2000 s channel 3 and 4 the response from the two calorimetry chip thermopiles Block 2 2002 time Time registered by the Keithley 2002 software data recorded block this is measured from an external clock and by the Keithley is consistent in all modules 2002 2002 volt Voltage measured by the Keithley 2002 the voltage difference between channels 3 and 4 Block 3 2400 time Time registered by the Keithley 2400 software data recorded block this is measured from an external clock and by the Keithely is consistent in all modules 2400 2400 volt Voltage supplied by the source meter a string of values of either O or 1 2400 curr Current measured by the Keithley 2400 in response on the 1 volt wave supplied to the chips 158 Table A 9 cont Legend for Figure A 34 Block 4 data recorded by the Lakeshore 331 Temperature Controller 33 time Time registered by the Lakeshore 331 Temperature Controller software block this is measured from an external clock and is consistent in all modules 331 setpoint Setpoint temperature set and registered by the Lakeshore 331 Temperature Controller 331 temp Sample space temperature registered by the Lakeshore 331 Temperature Controller 331 result Resistance value of the temperature depended carbon glass resistor used to monit
99. his is the first time that such a concrete permanent magnet potential has been observed in an Fe P type system Such an observation cannot be overlooked and it may now open new and exciting opportunities for novel permanent magnet alloys in this rich family of materials Further study and research is highly desirable 7 6 Conclusions Given their attractive mechanical properties a study on the Mn Fe P Ge and Mn Fe os P Ge systems was conducted with the purpose of making such systems economically viable for magnetic cooling applications The mapping of a limited compositional range of the Mn Fe P Ge system revealed a strong non linear Tc dependence with both Fe and Ge contents giving rise to the possibility to reduce the amount of expensive Ge by the increase of the Fe content Such a map also offers an understanding on the general behavior of Fe P type systems at high Fe concentrations The magnetic behavior of Fe rich Mn Fe P Ge is quite complex with the discovery of an unidentified magnetic phase at low Ge concentrations Unexpectedly the increase in Fe content also gives rise to a decrease in the Ferro Paramagnetic transition sharpness symptomatic with the disappearance of the first order behavior of this transition an unfortunate characteristic that makes this system unviable for MCE applications The Mn Fe os P Ge system maintains a sharp transition with a high magnetic moment at high Fe concentrations with the possib
100. hore 331 Temperature Controller are connected via a GPIB bus It is designated as Calorimetry vi and upon opening it one should be confronted with a panel such as Figure A 16 126 gt spt Application Font For er CO ota fu o 5435320000 0 00002722 aa 35921902 Figure A 16 Front panel of the interface software This interface is built up of 5 blocks one for each of the external measurement modules the four AMI modules are technically control modules not measurement modules clearly indentified on their headings and an extra one for data saving issues All relevant information given by the front panels of the modules during a measurement is also given by these blocks The Keithley 2000 block is displayed in detail in Figure A 17 with the respective legend given in Table A 3 2435920000 3 0 00D02722 Cm Figure A 17 Keithley 2000 interface software block 127 Table A 3 Legend for Figure A 17 1 On Off button 2 Indication of the channel currently being measured by the setup During a measurement these two displays will oscillate between 3 and 4 with a slight lag between them as one indicates the channel switching in the software and the other the channel switching in the hardware 3 From left to right time channel and voltage response from that same channel This except time should be equally displayed on the front panel of th
101. ial Since this point there has been a constant search for more effective cheap and non toxic working materials which may solidly establish the possibility for such a working refrigerator 1 3 Magnetocaloric materials The field of MCE material research drastically changed when in 1997 Pecharsky and Gschneidner 10 discovered what became known as the Giant Magnetocaloric Effect GMCE in GdsSi gt Ge This compound displays a strong magneto structural first order phase transition below room temperature which can be driven by temperature 11 magnetic field 12 and pressure 13 Such a transition is most interesting for MCE applications because it represents a sharp and radical change in the entropy of the system This result has effectively shaped and directed the research of new MCE materials We can currently name four particular families of compounds that are serious candidates for practical MCE applications the La Fe Si based compounds the MnAs based compounds the Heusler alloys and the Fe P based compounds 14 1 3 1 La Fe S1 5 based compounds Fe rich La Fe Si exhibits a magnetic ordering temperature between 200 and 260 K and has a sharp 1 order phase transition that can be manipulated by adequate heat treatments 14 These compounds can be further tuned by negative pressure expansion of the unit cell via insertion of hydrogen which can shift the transition temperature up to room temperature and beyond up until 450 K
102. ied transition around 372 K 78 6 4 Results on the Mn Co Si P system Early in this study it became obvious that this system did not in fact exist as a single phase material which might explain the limited number of compositions within this system studied by previous authors 11 14 All samples produced in this system even those belonging to the restricted Mn Co 3Si system displayed a large number of secondary phases This made it impossible to ascertain if any of these materials effectively belonged to the Mn Co Si P system The reason behind this difficulty to achieve a single phase sample even in compositions known to exist might be explained by the extremely rich and complex phase diagram of the Mn Co P system 24 demonstrated in Figure 6 5 The complex phase diagram makes this an extremely sensitive system to work with Mn Mn ht1 800 C 1073K MnossSio s rt 80 Mnos2Sisg MnsSi ht 70 VIN bN o Si MnsSis 30 10 20 30 40 50 60 70 80 90 Co Co rt CoSi at Si cosi CoSi Si Figure 6 5 The complex phase diagram of the Mn Co P system an insight into the difficulty of achieving a single phased sample in this system Figure taken from 25 79 6 4 1 Inverse magnetocaloric effect Samples produced with 0 1 P 0 5 and 1 2 Co 1 6 display multiple phases and exhibit a clear and significant magnetic transition from either an antiferromagnetic or ferrimagnetic to a ferromagnetic phase o
103. ility of lowering Ge content down to 0 12 and thereby effectively reducing the cost of this system to under half of the originally studied compositions Given the difficult balance of magnetic properties and element concentrations Tc cannot however be raised above 225 K but it is nonetheless easily tuned below this temperature in a region where there aren t many attractive magnetocaloric materials Remarkable and exciting properties were detected in a limited concentration range of Mn and Ge poor Mn Fe 95 P Ge which may indicate a tangible possibility 102 to use Fe P type alloys for permanent magnet applications This result points to a new and exciting field of permanent magnet research References 1 2 3 4 10 11 12 13 14 15 N T Trung Z Q Ou T J Gortenmulder O Tegus K H J Buschow and E Br ck Appl Phys Lett 94 102513 2009 N H Dung L Zhang Z Q Ou and E Br ck Appl Phys Lett 99 092511 2011 H Yabuta K Umeo T Takabatake K Koyama and K Watanabe J Phys Soc Jpn 75 2006 113703 H Yabuta K Umeo T Takabatake L Chen Y Uwatoko J Magn Magn Mater 310 2007 1826 A Yan K H M ller L Schultz and O Gutfleisch J Appl Phys 99 2006 08K903 W Dagula O Tegus B Fuquan L Zhang P Z Si M Zhang W S Zhang E Br ck F R de Boer and K H J Buschow IEEE T Magn 41 2005 2778 M Yue Z Q Li X B L
104. in most cases only X and K gt play a relevant role in the anisotropy energy density these three equations become equivalent to the second term development of Equation 2 13 In this case their solutions in terms of K and K naturally become the same and the following cases can be distinguished 1 For K K 0 the system is an isotropic ferromagnet 2 For K gt 0 and K K we have an easy axis for 6 0 usually z 3 For K gt 0 and K K or K 0 and K lt K 2 the plane perpendicular to the z axis is the easy plane 15 4 For 2K K 0 the easy axis will be reached for a O value given by the following expression 4 K 2 16 T 2 16 sin 0 making in fact an easy cone aligned with the z axis 3 o Cubic system The Cubic case is more complex given its high symmetry and it becomes easier to step outside of spherical coordinates In this case our expression becomes 2 fO lt Ka f BB L a Ko gy 2 17 with lt cos sinO p sin sinO and y cos In this case the anisotropy variations for different values of K and K are listed in Table 2 1 and 2 2 2 Table 2 1 Easy and hard axis for a Cubic system taking K 50 and x K K oo x 9 9 x 9 4 9 4 lt x lt 00 lt 100 gt Medium hard Easy Easy lt 110 gt Hard Hard Medium hard lt 111 gt Easy Medium Hard Hard Table 2 2 Easy and hard axis for a Cubic system taking K lt 0 and x K IK
105. in the base line of the magnetization curve indicating the increasing presence of secondary phases as shown in Figure 5 10 65 4 4 A gt 4 4 44 3 74 e q qto ti 4 B Li MU A EE It wet o e Normalized a e 0 95 15 20 25 30 35 40 45 50 55 60 Magnetization Am kg 40 MnFe SiB Temperature K 30 a x 0 4 4 x 0 02 207 x 0 05 m a A e 4 x 0 Emi aa 1 Tesla o 1 Tesla 0 50 100 150 200 250 300 350 400 Temperature K Figure 5 10 Magnetization vs temperature curves for MnFe SiB with x 0 0 02 0 05 0 1 and 0 15 demonstrating the effects of boron addition to the magnetic behavior of the cubic phase of Si rich Mn Fe Si P system Up to x 0 05 we can observe an increase in Ty magnetization and transition sharpness beyond this concentration all these properties lose their desirable characters Insert detail on the normalized antiferromagnetic ferromagentic transition demonstrating the increase in Ty up to x 0 05 and its subsequent decrease with x 0 1 and 0 15 In Figure 5 11 the dependence of Ty and the lattice parameter a on the added boron concentration is shown We were able to effectively increase Ty from 50 5 K for x 0 to 57 K for x 0 05 Further addition of B presented itself as counter productive in every aspect Even thought we were indeed able to optimize this transition we were
106. information Parallel to the previously described wires is the temperature control equipment This is divided into a small resistor inside the sample space see Figure A 10c and the sample space heater s The resistor and the heater s are connected to the Lakeshore 331 Temperature Controller on the outside of the setup The wiring details of this part of the setup are displayed in Figure A 12 with the respective legend given in Table A 2 Lakeshore 331 Temperature controller Sample space Temperature heater s dependent resistor Figure A 12 Wiring for the resistor and sample space heater s 123 Table A 2 Legend for Figure A 12 W G B and Y Color code for the 5 6 7 and 8 wires meaning Green White Black and Yellow Wires 5 6 7 Wires connecting the temperature dependent resistor to the Lakeshore and 8 331 Temperature Controller on the outside of the setup enabling for a four point temperature measurement hl and B wires Wires connecting the sample space heater s to the Lakeshore 331 Temperature Controller on the outside of the setup It should be relevant to note that the sample space heater s wiring is not directly related to the movable insert head as it is a parallel and semi independent system whose wires are merely bundled together with the ones from the insert head on the insert body Also one should not to confuse the chip heaters with the sample space heater s as these
107. ing potential for permanent magnet application in a certain low Mn and Ge concentrations Chapter 8 displays the microcalorimetry results on pure Fe P measured on the experimental setup described in Chapter 4 These demonstrate the gradual loss of the 1 order character of this compound s magnetic transition with increasing applied field arriving at a critical value of 0 4 T when the transition becomes completely 2 order Further temperature shift in this transition of about 1 5 K is observed at an applied field of 0 6 T 166 These results were in truth quite unexpected and could only be clarified if we consider that the sample measured is a near single crystal positioned in the equipment along its hard axis Further measurements are required in order to clarify this issue Finally the Addendum is a detailed step by step operator s manual for the experimental setup described in Chapter 4 and used in Chapter 8 This manual details all relevant aspects of this setup from hardware to software covering handling wiring sample preparation and measurement and data calculation 167 168 Samenvatting Gegeven de mogelijkheden van Fe P gebaseerde legeringen voor magnetocalorische toepassingen behandelt dit proefschrift verscheidene verwante materiaal systemen gerelateerd aan deze rijke familie van verbindingen zoals Mn Fe 3 Si P Mn Co 3 Si P en de Fe Co 3 Si P Eveneens behandeld is de opbouw van een microcalorimetrie opstelling bedoel
108. irst order phase transitions and giant magnetocaloric effect Delft University of Technology 2010 PhD thesis Chapter 3 Rodriguez Carvajal FULLPROF A Program for Rietveld Refinement and Pattern Matching Analysis Abstracts of the Satellite Meeting on Powder Diffraction of the XV Congress of the IUCr p 127 Toulouse France 1990 M McElfresh Fundamentals of Magnetism and Magnetic Measurements Featuring Quantum Design s Magnetic Property Measurement System printed by Quantum Design 1994 http hyperphysics phy astr gsu edu hbase solids squid html 31 7 8 32 H K D H Bhadeshia Differential scanning calorimetry University of Cambridge Material Science amp Metallurgy http www uzaktanegitimplatformu com UEP uep_ylisans ey2 ey2_download DSC 20Thermal2 pdf S Kasap D Tonchev Springer Handbook of Electronic and Photonic Materials Wemding Springer 2006 Chapter 19 Chapter 4 Differential microcalorimetry setup 4 1 Introduction As explained in Chapters 1 and 2 there are two common methods by which one may calculate the magnetocaloric effect of a given material the isothermal process in which we calculate the magnetic entropy change AS and the adiabatic process in which we calculate the actual temperature change AT of a given material Of these two the isothermal process is by far the most commonly used as in practical terms one only needs to perform a series of magnetic isothermal me
109. ith the exception of the plotting code lines grabtau m Clear all Parameters filename channel csv Sfirst column time second column thermopile voltage measurements pif filename switch csv Sfirst column time second column chip heater output in volts CSV seperator the seperator in the csv files 135 nSkip 8 number of data points to skip to get rid of the high slope nLength 100 number of data points the fit length is nMinimal 100 number of data points one curve should be atleast fitorder 1 2 works but 1 is better for theoretical and practical reasons use pif as switching time 1 0 or 1 delta bar 0 00001 some small value nGraphPre 20 on the worst fit graph how many data points to show before the fit nGraphPost 100 Son the worst fit graph how many data points to show after the fit tCycleTime 90 for use in the graphs amount of time one cycle takes show analysis 1 0 or 1 show cycle 1 0 or 1 remove old figures 1 0 or 1 seebeck 0 002060 V K for both ROUGH estimation approx power 0 00018 SW per chip for both Estimated Define data variables data dlmread filename csv seperator data time data 1 data value data 2 data length length data time data abs abs data value absolute value 136 data delta LU data abs 1 data length 1 data log
110. iu H Xu D M Liu J X Zhang J Alloy Compd 493 2010 22 M T Sougrati R P Hermann F Grandjean G J Long E Br ck O Tegus N T Trung and K H J Buschow J Phys Condens Mat 20 2008 475206 M Yue Z Q Li X L Wang D M Liu J X Zhang and X B Liu J Appl Phys 105 2009 07A915 M Yue Z Q Li H Xu Q Z Huang X B Liu D M Liu and J X Zhang J Appl Phys 107 2010 094939 D Liu M Yue J Zhang T M McQueen J W Lynn X Wang Y Chen J Li R J C Xubo L Z Altounian and Q Huang Phys Rev B 79 2009 014435 N H D ng Moment Formation and Giant Magnetocaloric Effect in Hexagonal Mn Fe P Si Compounds Delft University of Technology 2012 PhD Thesis Chapter 8 N T Trung First order phase transitions and giant magnetocaloric effect Delft University of Technology 2010 PhD thesis Chapter 4 E Br ck O Tegus L Zhang X W Li F R de Boer and K H J Buschow J Alloy Compd 383 2004 32 E Br ck J Phys D Appl Phys 38 2005 R381 103 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 104 O Tegus B Fuquan W Dagula L Zhang E Br ck P Z Sia F R de Boer K H J Buschow J Alloy Compd 396 2005 6 Tegusi Novel Materials for Magnetic Refrigeration University of Amsterdam 2003 PhD Thesis Chapter 5 J M D Coey J Magn Magn Mater 248 2002 441 J M D C
111. l compositional range of this system seems to be largely disregarded as most papers simply focus on the Co MnSi compound which displays quite promising properties for the already mentioned spintronics research 11 14 This particular compound is reported as crystallizing in the Fm3m cubic structure 11 13 15 with a lattice parameter of a 5 654 In terms of magnetic behavior it is reported to have a Tc of 985 K one of the highest among all known Heusler compounds 11 13 16 and a magnetic moment of 5 01 up f u On all accounts the P rich side of this system appears not to exist 15 with the only study done on it consisting of theoretical calculations meant for the same spintronic applications as the Si rich side 18 6 3 Results on the Fe Co Si P system 6 3 1 Si rich samples All samples produced in the Fe Co 3Si system were single phase and show the cubic Fm3m phase In the studied compositional range with Co concentration ranging from 2 to 1 2 the lattice parameter a was found to increase consistent with previous results described in section 6 2 1 In this same Co range the addition of elemental P caused the formation of a secondary phase with the orthorhombic Pnma structure demonstrated in Figure 6 7 The rise of this secondary phase is demonstrated in Figure 6 1a This phase becomes gradually more prominent with increasing P content while it never becomes the single phase in our samples This result indicates that it do
112. les on Xensor microcalorimetry chips and their subsequent fitting with the insert and 9 Tesla Cryostat At this point there will be no concern with actually measurement merely sample and chip handling 0 Should you already have two microcalorimetry chips mounted with a reference and measurement sample skip to point 7 1 Select two microcalorimetry chips from the reserve This should be done with a considerable degree of attention as a simple naked eye examination of a chip may be enough to detect a broken calorimetric membrane or disconnected wires see Figure A 23 Figure A 23 Close up of a XEN 39328 microcalorymetry chip evidencing the SiN membrane and chip wires 141 2 Take the two selected chips to a microscope and confirm their physical integrity 3 Fit both chip on the chip platform on the insert head and mount it on the insert At this point there is no need to mount any shields or heaters 4 On the Keithley 2000 press the bottom reading DCV making sure the REM definition is off and using the arrows check the values from channel 1 trough 4 they all should read 0 V with the occurrence of some minor noise While doing this also check the Keithley 2002 the numbers it displays are not important but should everything be as it should on no channels should you ever read an Overflow Press the 02 button and once again check channel 1 trough 4 You should read about 1 2 KQ on channels 1 and 2 and about
113. lft Leiden 2013 32 ISBN 978 90 8593 170 6 Printed by Proefschriftmaken nl Uitgeverij BOXPress Published by Uitgeverij BOXPress s Hertogenbosch dedicavit ad nihili Table of Contents Chapter 1 Introduction 1 1 Global refrigeration 1 2 Magnetocaloric effect 1 3 Magnetocaloric materials 1 3 1 La Fe Si 4 based compounds 1 3 2 MnAs based compounds 1 3 3 Heusler alloys 1 3 4 Fe P based compounds 1 4 Thesis outline References Chapter 2 Theoretical approach 2 Basic magnetocaloric quantities 2 2 Magnetic anisotropy 2 2 1 Magnetocrystalline anisotropy 2 2 2 Anisotropy energy density according to structure 2 2 3 Estimation of anisotropy constants References Chapter 3 Experimental equipment 3 Introduction 3 2 High energy planetary ball mill 3 3 Arc melting furnace 3 4 X ray diffraction 3 5 Scanning electron microanalysis and Energy dispersive X ray spectrometry 3 6 SQUID magnetometer 3 7 Differential scanning calorimeter References Chapter 4 Differential microcalorimetry setup 33 4 lIntroduction LLL 33 4 2 Experimental setup LL 34 4 2 1 Cryostatandinsert DA 4 2 2 Temperature control DO 4 2 3 Microcalorimetry chips 3 4 2 4 External measurement and power supply equipment 39 4 3 Measurement cycle on ee len 30 4 4 Data management 41 4 5 Conclusion Arn ete Behe dere een ear Breet hen rh ne 44 References o eeN EEN E E 44 Chapter 5 Magnetostructural study of the Mn Fe P Si system
114. librium materials and mechanical coating 1 In the current work this technique schematically represented in Figure 3 2 was mostly used as a pre alloying technique and homogenizer of starting materials turning our several elemental or binary initial compounds into an evenly homogenous powder which could be subsequently pressed and annealed All ball milled samples were produced in a Fritsch Pulverisette planetary mill and were milled for 6 hours 3 hours with 5 minute breaks every 5 minutes to prevent overheating at 360 rpm in 80 ml hardened steel crucibles each containing fifteen 4 g hardened steel balls and a sample mass of 5 g amounting to a sample Wall ratio of 0 083 3 Figure 3 1 a Arial diagram of a planetary ball mill b Sectional view of a planetary ball mill Legend angular moment of the planetary mill angular moment of the crucibles inside the mill R Radius of the planetary mill R radius of the crucibles inside the planetary mill Ra distance between the center of rotation of the mill and the crucible 1 3 3 Arc melting furnace Arc melting is widely used both in industry and lab scale applications given its suitability for the rapid production of highly homogeneous metallic ingots The used arc melting furnace shown in Figure 3 2 is a home made setup originally produced in the Van der Waals Zeeman Institute at the University of Amsterdam and later transferred to FAME This system was designed for
115. lue of H that makes Ms parallel to the applied field is reached when sin 0 1 the anisotropy field H y can be calculated as _ 2K 4K MH an 2 20 S Given that in some materials K is negligible measurements of Hay are then sufficient to determine K Sucksmith Thompson method Still alternative methods for the determination of K and K have been developed One such method developed by Sucksmith and Thompson in 1954 8 is based on the following relation with J 4M P f 2 21 S S which is valid for magnetization curves of single crystals under small fields perpendicular to the easy direction It is then possible to assume that Js does not change with the field strength and that sin 9 J J which when substituted in Equation 2 19 give us Equation 2 21 When H J is plotted versus J the anisotropy constant K can be estimated by the vertical interception of the graph with the y axis and K by the slope of this same graph e Modified Sucksmith Thompson method Another method based on the Sucksmith Thomson has also been proposed by Ram and Gaut in 1983 9 which has the advantage of weeding out errors which occur when this method is used on powder samples with a misalignment In this modified model H a J J is plotted versus a J J J J Y where J is the remanence in the hard direction and J J J has been introduced to simulate a perfect magnetic alignment of the powder particles
116. lways a pleasure Of course one cannot forget the many professionals and technicians of FAME who were always willing to give a hand when time permited it Fred I probably bothered you more than anybody else and yet unlike Anton who tried to physically run away you were always patient courteous and well disposed Much like Peter I owe you a great deal you are a professional of extraordinary skill and I hold you in the highest possible regard Jouke you were always surprisingly available for any and all 174 informatics problems anyone had and even though I don t think this was ever in your job description you always went out of your way to help in anyway you could Michel I never got this why do you sit on a giant inflatable ball Kees it was nice to know that somebody else thought the ICDD were a bunch of lazy bums Paul you are a gentleman our hard working secretaries Nicole and Ilse a big thank you Besides all of these there was of course a constant stream of Bachelor and Master student passing by this place that helped easy the heavy work days Naming them all would be an ungrateful task but I would still like to take the time to remember a hand full of which I was particularly found of Chris van Soest whatever it is that makes the average Dutch youngster into a jackass you managed to make that into something cool Cassandra flos sanctorum Hanan you are a bright and sweet person and don t think you don t have what it t
117. m If the problem is with the values in the temperature controller you most likely have a wiring problem as the thermometer itself is just a simple resistor These issues will not be dealt with in the current manual as they boil down to adequately using a hand held multimeter and wire soldering 6 If all seems well dismount the insert head and carefully remove both chips Take the chips to a balance and take the following steps with one at a time 1 Weight the chip take note of its mass and tare the balance to 0 2 With a tip of a thin tweezer take an excitingly small amount of grease Apiezon N if you re planning on doing a low temperature measurements and Apiezon H for high temperature measurements and slightly touch the SiN calorimetry membrane so as a small droplet is left on its surface Should you notice that during this process some grease has extended beyond the membrane and into the chip frame then this chip has become inadequate go back to point 1 and follow this procedure with one chip Should you see that during this process the membrane has broken as a result of over pressure from the tweezer then this chip has become inadequate go back to point 1 and follow this procedure with one chip 3 Weight the chip with the grease take note of its mass and tare the balance to 0 An ideal grease amount should be so small as the balance will not be able to register it 4 With a thin tweezer transport a small particle of sample
118. mmm Pmmm Pmmm P6 mmm Pmmm P6 mmm I4 P6 mmm 4 P6 mmm Pmmm Pmmm 4 Pmmm I 4 Pmmm 4 PmmmH 4 I4 Pmmm I4 Pmmm I4 Antiferro Para Antiferro Para Antiferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Ferro Para Antiferro Para Ferro Para Ferro Para Ferro Para Ferro Para Antiferro Para Antiferro Para Antiferro Para Antiferro Para Antiferro Para Antiferro Para Antiferro Para Antiferro Para Ferro Para Ferro Para Ferro Para 111 100 91 107 95 85 97 108 125 127 164 221 667 697 103 127 84 71 75 84 88 133 117 133 78 154 233 223 213 523 695 61 MnooFe gt 1Sio1Poo PmmmH 4 Antiferro Para 227 MnosFe gt gt Sio1Poo Pmmm 4 Antiferro Para 219 Mno Fe23Si0 1Po9 I4 Pmmm Ferro Para Mno 6Fe24Sio 1Po9 14 Pmmm Ferro Para Mn P 14 Antiferro Para 150 Mn sFeo sP 14 Antiferro Para 117 Mn FeP I4 Pmmm Antiferro Para 89 Mn Fe sP Pmmm Antiferro Para 233 MnFe P PmmmH 4 Antiferro Para 240 Mng Fez P PmmmH 4 Ferro Para 400 Mno gFe22P PmmmH 4 Ferro Para 400 Mn Fez 3P I4 Pmmm Ferro Para 413 Nnn abe 4P I4 Pmmm Ferro Para 42 Mna spe 5P I4 Pmmm Ferro Para 473 Fe P Ia Ferro Para 686 5 6 Mn Fe 3 Si P B compounds In light of the results obtained during the study of the Mn Fe 3 Si P system there was a further interest in exploring the antiferromagnetic ferromagnetic transition obs
119. n Figure 6 8 makes this compound an extremely exciting candidate for new inverse magnetocaloric effect systems 75 m Mn Co Si 1 Tesla TaT 60 45 30 15 Magnetization Am kg 0 50 100 150 200 250 300 350 Temperature K Figure 6 8 Magnetization vs temperature curve for the Mn Co sSio gt Pos compound demonstrating an extremely high inverse magnetic transition 6 5 Conclusions The initial motivation to explore the magnetocaloric potential of the novel P6 mmm hexagonal phase discovered in the Mn Fe Si P system Chapter 5 led to a systematic study of the magnetic and structural properties of the Fe Co 3 Si P and Mn Co Si P systems Our results on the Fe Co Si P system strongly suggest that this quaternary system does not exist as a single phase compound While we were very successful in producing samples belonging to the ternary Fe Co 3Si and Fe Co P systems the addition of P or Si to these always gave rise to multi phase samples This result 82 motivated us to further explore methods to tune Tc in the Fe Co 3P system that did not resort to Si substitution leading us to a promising and unexpected result with the S substitution instead All samples belonging to this new Fe Co P S system have a slightly lower Te than those not containing S and showed a new and unexpected secondary transition around 370 K Similar to the Fe Co Si P system the Mn Co Si P system d
120. n magnetization 2 strong magneto crystalline anisotropy with a uniaxial crystal structure 3 a high Tc 26 and 4 a high coercivity field such a result naturally becomes relevant Pure Fe P is in fact currently studied as a possible alternative to the current rare earth based magnets given that it is a cheap compound which exhibits a considerable magnetic anisotropy 25 27 This property however is not sufficient for an effective application as a significant drawback of this material is its low Tc about 217 K 28 and attempts have been made to raise this value by elemental substitution such as P by Si This however frequently results in the loss of the hexagonal P 6 2m structure in favor of the orthorhombic mm2 structure 27 as shown in Figure 7 10 98 C 7 b Fe Si P Figure 7 10 Imm2 orthorhombic structure found for the Fe Si P system The mentioned MnjFe Po9Geo sample which contains small amounts of both Mn and Ge was found to have a high Tc while still maintaining the desired P6 2m hexagonal structure Such a combination of factors indeed changes the outlook on Fe P based permanent magnets as for the first time we have a reliable direction in which to explore the possibility of future magnetic applications In order to further study this phenomenon two different types of magnetization versus field measurements were performed All samples that display this discrepancy between FC and ZFC measurement
121. n of the novel hexagonal phase discovered in Chapter 5 in an attempt to use and tune it for MCE applications namely through the addition of Co resulting in the Mn Co Si P and Fe Co Si P systems Although the results from this research did not provide us with viable MCE materials it none the less opened the possibility for further research in the Fe Co 3 P S system and the Mn Co 9 Si P system which displays a substantial inverse MCE Chapter 7 focuses on the development of Fe P type alloys and consists of a study of the Mn Fe P Ge system to optimize it in terms of monetary costs This also led to the study of this system s potential for permanent magnet applications Chapter 8 is a fundamental investigation of the key magnetic properties of pure Fe P using the experimental setup described in Chapter 4 namely the change in the nature of the magnetoelastic transition of this system when put under a magnetic field Finally a detailed Addendum is included in this thesis consisting of an operator s manual for the experimental setup described in Chapter 4 and used in Chapter 8 References 1 B Nagengast Mech Eng Mag May 2000 56 2 W T Tsai Chemosphere 61 2005 1539 3 R P Fernandes Magnetocaloric effect of Pr Ni Co hard magnets and Ni Mn Ga Bi shape memory alloys Universidade de Aveiro 2007 MSc thesis Chapter 1 4 K A Gschneidner and V K Pecharsky Annu Rev Mater Sci 30 2000 387 5
122. nel volt 2000 volt 2002 time 72002 volt 2400 time 2400 volt 331 setpoint 331 setpoint 331 temp 331 result 331 temp 331 result 33 1 heater 331 heater Save stop Le X axis Format Y Axis 2000 volt Channels x 2000 time 2000 channel 2000 volt 2400 curr 331 time 331 setpoint 331 temp 331 result 331 heater X Axis x Channels y 2000 time 2000 channel 2000 volt 2002 time 2002 volt 2400 time 2400 volt 2400 curr 331 time 331 setpoint 331 temp 331 result 331 heater Bi Y Axis 2000 time 2000 volt Figure A 34 Detail on the X and Y columns on the Calorimeter read vi Figure A 33 Location of the functions mentioned in the above instruction for the Calorimeter Read vi 157 4 As mentioned above section A 3 2 the Channel X and Channel Y columns list the data contained in the tdms file produced by the Calorimetry vi during a measurement originating from the various measurements performed by the several modules A detail of these columns is given in Figure A 34 with the respective legend given in Table A 9 Table A 9 Legend for Figure A 34 Block 1 2000 time Time registered by the Keithley 2000 software data recorded block this is measured from an external clock and by the Keithley is consistent in all modules 2000 2000 channel Registry of the channel in w
123. ng no particular directional preference Figure 2 1 Spherical free energy surface for an isotropical exchange 2 Assuming now that f m is defined as an expansion that only depends on fay m K K sin O K sin 0 K sin 0 2 13 where K K and K are anisotropy constants independent of m having the dimensions of energy per unit of volume For different values of these constants we obtain different energy surfaces with depressions indicating energetically favored directions as displayed in Figure 2 2 12 Figure 2 2 Broken spherical symmetry with the formation of an easy magnetization axis 2 The direction with the lowest value for fay m is referred to as easy magnetization direction and should this be coincident with a particular crystallographic axis it may be referred to as easy magnetization axis or simply easy axis This represents the direction in which magnetization will naturally align with in order to minimize the system s free energy and upon measuring the same magnetic sample for different orientations a distinct magnetic behavior will be recorded depending on the orientation of the sample relatively to the applied magnetic field as shown in Figure 2 3 1400 0001 M emu cm 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10 000 H Oe Figure 2 3 Example of magnetization curves measured along the easy direction 0001 demonstrating a rapid rise to the saturation m
124. nitially the sample and the chip are at the same temperature T A current is then applied to the chip s heater making the sample temperature rise to a new constant value 7 Next the current is cut and the sample relaxes back to its original temperature in a relaxation time 7 22 This cycle is achieved by supplying the chip heaters with a square wave from the Keithley 2400 and then via the chips 39 inbuilt thermopiles measuring the temperature of the sample and reference sequentially with the Keithley 2000 To calculate one data point we need to measure both chips Figure 4 5 shows the typical response for one relaxation period 6 cO A C Thermopile excitation mV 1355 1360 1365 1370 1375 1380 1385 1390 Time s Figure 4 5 Thermopile response to the square wave heater input used to perform a relaxation measurement The cooling curve has the following form V t Ae 4 1 The mentioned relaxation time is then calculated by taking the logarithm of the signal and performing a least square fit for a straight line The slope of this fit is equal to the inverse of the relaxation time 7 We can then calculate the heat capacity by the following simple expression C kt 4 2 40 where x the thermal conductivity 4 3 and P is the power supplied to the chip 4 4 Data management A complete measurement taken with this set up is depicted in Figure 4 6 6 Thermopile excitation mV
125. nto the insert head connect the second end of the current divider to it and connect it to the heater wiring coming from the insert body The current divider should not be dismissed as it has been determined that one single heater cannot withstand the High setting of the Lakeshore 331 Temperature Controller and should you want to make a measurement above 80 K you cannot avoid using the High setting 11 Repeat point 4 and 5 In the Lakeshore 331 Temperature Controller press the Setpoint button making sure the Remote definition is off and insert a setpoint which is higher then the sample space temperature reading Press the Heater Range button and select one of the power ranges of the heater using the arrow buttons restrain from using the High setting press Enter At this point you should read a percentage value under the 144 resistance value of the carbon glass temperature dependent resistor this is the percentage of the maximum output of the heater setting you have selected and that the Lakeshore 331 Temperature Controller is supplying to the heaters in order to raise the temperature of the setup to your selected setpoint If you see this it means your heaters are working accordingly and you can press the Heater off button If instead you read Open you have an error Make sure the heater wires are correctly connected to the back of the Lakeshore 331 Temperature Controller If all seems well dismount the t
126. oes not seem to exist as a single phase compound Even the compositions previously studied by other groups proved to be extremely difficult to produce as single phase samples The detection of an inverse magnetocaloric effect in some of these multiple phase samples led to the discovery of a new system with remarkable properties for inverse magnetocaloric effect applications showing a high magnetic transition temperature for an antiferromagnetic or ferrimagnetic to ferromagnetic phase or alternatively from an antiferromagnetic to a ferrimagnetic phase In both of these systems the hexagonal P vnmm phase was not found References 1 S Wurmehl G H Fecher H C Kandpal V Ksenofontov C Felser and Hong Ji Lin Appl Phys Lett 88 2006 032503 2 K Hamaya N Hashimoto S Oki S Yamada M Miyao and T Kimura Phys Rev B 85 2012 100404 3 S Wurmehl G H Fecher V Ksenofontov F Casper U Stumm C Felser Hong Ji Lin and Y Hwu J Appl Phys 99 2006 08J103 4 V Niculescu T J Burch K Raj and J J Budnick J Magn Magn Mater 5 1977 60 5 V Niculescu J I Budnick W A Hines K Raj S Pickart and S Skalski Phys Rev B 19 1979 453 6 L Hongzhi Z Zhiyong M Li X Shifeng L Heyan Q Jingping L Yangxian and W Guangheng J Phys D Appl Phys 40 2007 7121 7 Y Nishino Shin ya Inoue S Asano and N Kawamiya Phys Rev B 48 1993 13607 8 M Goto H Tange T Tokuna
127. oey Scripta Mater 67 2012 524 Z Zhang X Song Y Qiao W Xu J Zhang M Seyring and M Rettenmayr Nanoscale 5 2013 2279 J E Gould IEEE T Magn 5 1969 812 S Sugimoto J Phys D Appl Phys 44 2011 064001 M Sagawa S Fujimura N Togawa H Yamamoto and Y Matsuura J Appl Phys 55 1984 2083 J M D Coey J Magn Magn Mater 140 1995 1041 M Costa O Gr n s A Bergman P Venezuela P Nordblad M Klintenberg and O Eriksson Phys Rev B 86 2012 085125 S J Collocott J B Dunlop H C Lovatt and V S Ramsden Mater Sci Forum 315 317 1999 77 L Severin L H ggstr m L Nordstr m Y Andersson and B Johansson J Phys Condens Mat 7 1995 185 O Beckman L Lundgren P Nordblad P Svedlindh A T rne Y Andersson and S Rundqvist Phys Scripta 25 1982 679 B K Srivastava T Ericsson L H ggstr m H R Verma Y Andersson and S Rundqvist J Phys C Solid State 20 1987 463 R Fruchart A Rogers and J P Senateur J Appl Phys 40 1969 1250 Chapter 8 In field microcalorimety measurements on polycrystalline Fe P 8 1 Introduction As exemplified by the previous Chapters the core theme of the material research performed on magnetocaloric materials in Delft is focused on Fe P type alloys 1 7 The remarkable magnetocaloric potential found in these compounds is linked to the realization that they maintain many of the same properties as the parent
128. omiyoshi and H Watanabe J Phys Soc Jpn 39 1975 295 C Pfleiderer J Boeuf and H v L hneysen Phys Rev B 65 2002 172404 C Pfleiderer Physica B 329 333 2003 1085 A Go M Pugaczowa Michalska and L Dobrzynski Eur Phys J B 59 2007 I E J D Garba and R L Jacobs J Phys F Met Phys 16 1986 1485 G D Mukherjee C Bansal and A Chatterjee Physica B 254 1998 223 S Yoon and J G Booth Phys Lett 48A 1974 381 S Yoon and J G Booth J Phys F Metal Phys 7 1977 1079 V A Niculescu T J Burch and J I Budnick J Magn Magn Mater 39 1983 223 267 H Miki K Ohoyama H Onodera H Yamauchi Y Yamaguchi H Nojiri M Motokawa S Funahashi and S Tomiyosh Physica B 237 238 1997 465 466 P Mohn and E Supanetz Philos Mag B 78 1998 629 16 17 18 19 20 21 22 23 24 25 O A Dubovskii A V Orlov and V A Semenov Phys Solid State 45 2003 326 Rundqvist Acta Chem Scand 16 1962 992 S Rundqvist Acta Chem Scand 16 1962 1 A Broddefalk P James Hui Ping Liu B Kalska Y Andersson P Granberg P Nordblad L H ggstr m and O Eriksson Phys Rev B 61 2001 413 Hui ping Liu Y Andersson P James D Satula B Kalska L H ggstr m O Eriksson A Broddefalk and P Nordblad J Magn Magn Mater 256 2003 LIT M Goto H Tange T Tokunaga H Fujii and T Okamoto Jpn J Appl Phys 16 1977 2175 T
129. on junction 5 b Schematic representation of a the insert of a MPMS magnetometer c Schematic repetition of a detection coil of a MPMS magnetometer 6 3 7 Differential scanning calorimeter Differential scanning calorimeter DSC measurements were used on those samples whose transition temperatures exceeded the temperature range of our magnetometers For this end a Q2000 model from TA Instruments Waters LLC was used performing temperature sweeps from 0 C to 500 C at a rate of 20 C per minute The functioning of this equipment relies on a sample and reference assembly connected by a low resistance heat flow path a metal disc enclosed in a single furnace Any enthalpy or heat capacity change in the sample as those which happen during a structural or magnetic transition causes a difference in its temperature relatively to the 30 reference which is recorded using a calibration experiment 7 3 A diagram of this DSC system is given in Figure 3 6 Thermo Chromel electric disk constantan Thermo disk couple Heating junction block Figure 3 6 Schematic representation of a DSC system as used in the current thesis S stands for Sample and R for Reference 8 References 1 2 Liang Hao Yun Lu Hiromasa Sato Hiroshi Asanuma Jie Guo Int J Miner Process 212 2013 51 Pham Duc Thang Permanent Magnets based on Iron Platinum alloys University of Amsterdam 2003 PhD thesis Chapter 2 N T Trung F
130. on the magnetization curve of the aniferromagnetic orthorhombic samples revealing the characteristic bump of an antiferromagnetic transition The temperature induced transition from antiferromagnetism to canted ferromagnetism in compounds on the Si rich side of the structure diagram has been found to be completely independent of any compositional or magnetic field change maintaining a relatively constant temperature of about 50 K between a Mn content of 1 8 and 1 Below this composition range it rapidly decreases until it is no longer observable at a Mn content of 0 6 a result also observed by Yoon et al 11 12 The addition of P has proven to reduce the overall magnetic moment of the sample and widen the already broad second order phase transition between the ferromagnetic and paramagnetic phases resulting in an increase in Tc as seen on Figure 5 5 This effect is probably due to the increase in the lattice parameters by P substitution 55 m Mn Fe Si A Mn Fe Si P 0 8 0 2 N Co N A 0 1 Tesla Magnetization Am kg o N o D A e 0 50 100 150 200 250 Temperature K Figure 5 5 Magnetization versus temperature plot reveling the influence of P addition to the canted ferromagnetic phase of the Mn Fe Si P system An analysis of the temperature dependence of the magnetization for Mn Fe Si samples with 2 x 2 5 revealed an unusual magnetic behavior Instead of a continuous incr
131. or the temperature inside the sample space 331 heater Heater output supplied and measured by the Lakeshore 331 Temperature Controller With resource to this program you can select and visualize any two sets of data by selecting one as the X and other as the Y axis and pressing the button Read which will plot the two data sets in the large display 5 To continue with you data handling you will need to extract the following sets of data 1 X AXIS Y Axis 2000 time 2000 volt The chip voltage response measured by the Keithley 2000 vs time 2 X AXIS 2400 time Y Axis 2400 volt The voltage supplied by the Keithley 2400 Source meter vs time 159 3 X Axis Y Axis 331 time 331 temp The sample space temperature measured by the Lakeshore 331 Temperature Controller vs time In order to extract these data sets from the general meas tdms file while they are on display press the Ok button below Save Attention should be paid to the Output file box as the file produced should have the termination csv These functions are shown in Figure A 35 E el e wi alt Input File Path Save File D SharedData calorimeter Peter Vose Fe2P 16 11 12 meas tdms Gel Channels x Channels y HE 2000 time 2000 time 2000 channel F 2000 channel 2000 volt 2000 volt 2002 time 2002 time 2002 volt 2002 volt 240
132. ould be equally displayed on the front panel of the Lakeshore 331 Temperature Controller Resistance value of the carbon glass temperature dependent resistor inside the sample space This value indicates the temperature at which the sample space is at via the calibration curve presented in Figure 4 3 This should be equally displayed on the front panel of the Lakeshore 331 Temperature Controller Heater output in percentage This should be equally displayed on the front panel of the Lakeshore 331 Temperature Controller Remote setpoint temperature setter This allows you to remotely enter a new setpoint temperature value into the Lakeshore 331 Temperature Controller during a measurement 130 Table A 5 cont Legend for Figure A 19 8 Remote Heater range setter This allows for a remote control the heater setting of the Lakeshore 331 Temperature Controller 0 Off 1 Low 2 Medium 3 High during a measurement 9 Measurement frequency This it is present in every equipment block 10 Wellness indicators lighting either in green yellow or red Under normal circumstances these would light green otherwise they indicate an error These are present in every equipment block 11 Saving light This is an indicator of when you are recording your data The Keithley 2400 block is displayed in Figure A 20 with the respective legend given in Table A 6 Keithley 2400 nm gH uH fi
133. power acting as both sensors and actuators In this way the system is able achieve strict isothermal conditions The drawback of this system is a high dependence on an accurate calibration of the Peltier cells The versatility of Peltier elements can also be observed in the setup described by Porcari et al 7 which resorts to a power Peltier to perform the temperature control and two Peltier sensors to perform the actual DSC measurement 8 a setup quite similar to the one described by Jeppesen et al 9 Under this perspective the microcalorimetry chips from the company Xensor Integration have gained increasing relevance due to their precision practicality and relatively small price as presented by Morrison et al 10 11 Minakov et al 12 15 and Merzlyakov in a non magnetocaloric context 16 making them a very attractive and promising component for such calorimeters We report the design and construction of an experimental setup that allows for specific heat measurements under high magnetic fields using these microcalorimetry chips We have adopted a two chip setup in our equipment which enables us to easily bypass many bothersome calibration and equipment specific issues This instrument s potential range is well beyond the purely magnetocaloric oriented as it can provide invaluable information regarding any phase transition where the application of a magnetic field may play a significant role 4 2 Experimental setup 4 2 Cr
134. pplied field of 0 3 Tesla measured in this setup The reference used was a Cu sample and its specific heat value was taken For this measurement the used reference was a Cu piece and the measured sample polycrystalline Fe P this same sample is fully examined in Chapter 8 Their details can be found on Table 4 2 Table 4 2 Details regarding the samples in the two chips used in the measurement displayed in Figure 4 6 Mass mg Error mg Grease mg Error mg Sample 0 22 0 02 Grease mass was too small to be Reference 0 23 0 02 registered by our balance none the less we should consider a maximum value of 0 02 mg the balance error 43 4 5 Conclusion A setup to measure the heat capacity of micrograms of sample has successfully been assembled using microcalorimetry chips from the company Xensor Integration With this calorimeter it is possible to measure the specific heat capacities of samples in a magnetic field at various temperatures For this end we use the relaxation method where the heat capacity is calculated from the sample s relaxation time after the input of a square wave to the chip s heaters This setup uses two microcalorimetry chips simultaneously to eliminate the need for thermopile calibration and concerns regarding the Seebeck coefficient in magnetic fields References 1 2 3 4 5 10 44 A M Tishin and Y I Spichkin The Magnetocaloric Effect and Its Applica
135. properties and assemble a magnetostructural phase diagram for the Mn Fe 3 P Si system which is greatly valuable for its understanding and may yet open further horizons to those areas where both the Mn Fe Si and Mn Fe P systems have been studied in the past Even thought several of its compositional areas presented significant magnetization shifts none of these were sufficiently large or sharp to make this system appropriate for magnetic cooling purposes 5 2 Material overview 5 2 1 The Mn Fe Si system The Mn Fe Si system has been studied both for its properties as an itinerant electron antiferromagnet and as a half metallic ferromagnet HMF at different compositional ranges In its antiferromagnetic phase Mn rich it has been studied in the context of spin wave excitation by means of neutron inelastic scattering 3 As a HMF Fe rich having a band gap in one spin at the Fermi level whereas the other spin is strongly metallic a complete spin polarization of the conduction electrons at the Fermi level is found which is of interest in the field of spintronics 4 The MnsSi compound crystallizes in the cubic Fm3m structure AlFe prototype 3 5 6 It is a bee structure with two distinct Mn atom sites A Mnl and B B MnII 3 5 as shown in Figure 5 2a The reported magnetic moments are 0 19 ug and 1 7 Up 3 5 7 for Mnl and Mnll respectively This compound orders antiferromagnetically below a N el Temperature Ty of
136. pta 17 1978 39 L Caron M Hudl V H glin N H Dung C P Gomez M Sahlberg E Br ck Y Andersson and P Nordblad Phys Rev B to be published S Rundqvist and F Jellinek Acta Chem Scand 13 1959 425 D Scheerlinck and E Legrand Solid State Commun 25 1978 181 J Tobola M Bacmann D Fruchart S Kaprzyk A Koumina S Niziol J L Soubeyroux P Wolfers and R Zach J Magn Magn Mater 157 1996 708 M Costa O Gr n s A Bergman P Venezuela P Nordblad M Klintenberg and O Eriksson Phys Rev B 86 2012 085125 L Severin L H ggstr m L Nordstr m Y Andersson and B Johansson J Phys Condens Matter 7 1995 185 113 16 17 18 19 20 21 22 23 114 H P Scott B Kiefer C D Martin N Boateng M R Frank and Y Meng High Pressure Res 23 2008 375 J P Senateur A Rouault R Fruchart J J Capponi M Perroux Mat Res Bull 2 1976 631 R Fruchart A Rogers and J P Senateur J Appl Phys 40 1969 1250 O Beckman L Lundgren P Nordblad P Svedlindh A T rne Y Andersson and S Rundqvist Phys Scripta 25 1982 679 B Carlsson M G lin and S Rundqvist J Solid State Chem 8 1973 57 N Dass J Phys Soc Japan 28 1970 251 M McElfresh Fundamentals of Magnetism and Magnetic Measurements Featuring Quantum Design s Magnetic Property Measurement System printed by Quantum Design 1994 A A Minako
137. question increases reduces or an isothermal entropy change where the material remains at the same temperature but exchanges heat with its surrounding environment These two processes are demonstrated in Figure 1 2 BE d 4H Adiabatic p Process LQ Q 3 T 1 Adiabatic Process P 4 AT T AT C Magnetic i T refrigeration Vapor cycle refrigeration Figure 1 1 Comparison between a magnetic refrigeration cycle left and a gas compression cycle right Adiabatic process Isothermal process Temperature Figure 1 2 a Example of an adiabatic temperature variation with the application of a magnetic field Ericsson cycle b Example of an isothermal heat exchange with the application of magnetic field Brayton cycle c Example of the adiabatic and isothermal processes between two isofield entropy curves The calculation of these two quantities isothermal entropy change and adiabatic temperature change is described in detail in Chapter 2 The idea of a thermo magnetic refrigerator first appeared in the late 1920s when cooling via adiabatic demagnetization was proposed by Debye 6 and Giauque 7 The process was latter demonstrated by Giauque and MacDougall in 1933 where the remarkable temperature of 250 mK was reached 8 This was still a long way from an actual commercially viable refrigerator which was first proposed by Brown in 1976 9 with a refrigeration model using Gd as a cooling mater
138. r slit Secondary monochromator Nubi Sample Figure 3 3 Schematic representation of an X ray diffractometer image from PANalytical B V 3 5 Scanning electron microanalysis and Energy dispersive X ray spectrometry The determination of phase compositions in our Mn Co Si P samples Chapter 6 was conducted with a scanning electron microscope SEM and an energy dispersive X ray spectrometer EDS Both these measurements were performed in a Jeol JSM 840A equipment at the Delft Aerospace Structures amp Materials Laboratory at the Deft University of Technology The equipment in question had an acceleration voltage of 5 to 35 kV a magnification of 20 to 300000x and a resolution down to 3 5 nm A schematic representation of the EMP and EDS setup is given in Figure 3 4 28 filament second objective lens LN2 dewar viewing microscope SQ crystal electron detectors a isolation valve detector 9 diffusion pump ES sample current microammeter to backing pump Figure 3 4 Representation of a generic EMP EDS image taken from the Department of Geosciences of the University of Wisconsin Madison 3 6 SQUID magnetometer Superconducting quantum interference devices SQUID are usually used for the precise measurement of magnetic moments This equipment resorts to a sensor ring consisting of two superconductors separated by thin insulating layers to form two parallel Josephson junctions sc
139. r a linear extrapolation an estimated value of around 49 6 Am kg should be expected for Mno Fe23Sio 2Po 3 Instead we find a value of 17 52 Am kg showing that this sample is now predominantly antiferromagentic Through simple calculus we can then assume that our Mno Fes 2510 2P s sample is made up of about 35 of ferromagnetic tetragonal phase and 65 antiferromagnetic orthorhombic phase This method of phase estimation however useful can only be used on borders between structures with a distinctly different magnetic behavior 54 AA AAAA AAM AAA MA A AM A 10 28 MMA MAM AMA A AA A mv PPAR Ee rro tetra 10 2 8 9 9 6 300 a Temperature i Antiferro ortho Magnetization Am2 kg FEARS AA A BELERELE AURORA P e pa pp T 21 2 amp Ferro tetr AAA lagnetization Am kg EC EE Antiferro ortho Ferro tetra Mn Fe Si 23 02 Pos h Mn Fe Si P 24 02 0 8 Magnetization Am kg o gt A Magnetization Am2 kg 250 Temperature i Antiferro ortho 0 50 100 150 200 250 300 350 400 Temperature K 30 25 20 60 50 m 40 30 20 Figure 5 4 Magnetization versus temperature plots revealing the influence of Si content on the magnetic behaviors of the compositions on both sides of the border between the orthorhombic and tetragonal phases taken with 1 Tesla of applied magnetic field Inserts detail
140. r alternatively from antiferromagnetic to a ferrimagnetic phase As shown in Figure 6 6 this leads to an increase of magnetization for increasing temperature resulting in an inverse magnetocaloric effect m x 1 4 y 0 1 4 X 1 3 y 0 3 vw x 1 3 y 0 5 0 5 Magnetization Am kg 0 50 100 150 200 250 300 350 400 Temperature K Figure 6 6 Example of some of the produced samples displaying an inverse magnetocaloric effect The non linear evolution of both transition temperature and overall magnetization with the change in both the MniCo and Si P ratios is symptomatic of the apparently random occurrence of secondary phases in this system of compounds In this process contrary to the usual magnetocaloric effect where a material heats up with the adiabatic application of an external magnetic field a material cools down by the application of a magnetic field 26 28 This effect is usually observed in magnetic transitions between antiferro ferromagnetic collinear and non collinear antiferromagnetic or antiferro ferrimagnetic phases 26 80 In terms of applications this effect is most often considered as a possible heat sink for a functioning magnetic refrigeration device 29 30 meaning that when a field is applied to a regular magnetocaloric material the heat released can be effectively dumped into the inverse magnetocaloric material which cools down with the application of the same field Alternatively a magnetoc
141. r the 9 Tesla superconducting magnet represented in Figure A 7 Controlled by the Model 430 Power Supply Programmer Figure A 7 Model 4Q0625PS Model 185 liquid nitrogen meter used to monitor the liquid nitrogen level inside the AMI cryostat represented in Figure A 8 I um l Dodot 108 doque Lavet drm een t e sls 9 Figure A 8 Model 185 Liquid Nitrogen Level Monitor Model 135 liquid helium meter used to monitor the liquid helium level inside the AMI cryostat represented in Figure A 9 119 POWER SENSOR OH CURRENT cs AMI N 9 0 Lever Model 135 Liquid Helium Level Monitor RAISE Lo SETPOINT SET LENGTH INT INTERVAL SILENCE Figure A 9 Model 135 Liquid Helium Level Monitor A 2 Wiring The understanding of the relation between the inside and outside of the setup is entirely dependent on the wiring of the whole microcalorimetry instrument as this runs from the very tip of the movable insert up to the measurement and control modules There are several wiring clusters in the setup that should be understood 1 The movable insert head 2 The insert body 3 The 16 pin plugs on the top of the insert 4 The 16 pin cable plugs on the outside of the setup 5 The wires that finally immerge from the cables and feed into our measuring equipment on the outside of the setup A 2 1 Insert head This should be regarded as the heart of the
142. reference to the chip membrane and weight the chip 5 Do this as many times as required until you have the desired mass on the chip betewn 0 2 to 0 8 mg Take note of this mass 143 Should you see that during this process some grease has extended beyond the membrane and into the chip frame then this chip has become inadequate go back to point 1 and follow this procedure with one chip Should you see that during this process the membrane has broken as a result of over pressure from the tweezer then this chip has become inadequate go back to point 1 and follow this procedure with one chip 7 After both chips are ready take them to a microscope to determine with certainty if any grease has extended beyond the membrane and into the chip frame or if the SiN has broken as a result of over pressure from the tweezer If you see that this has happened go back to point 1 and follow this procedure with one chip 8 Once again place the two chips in the chip platform in the insert head and make a simple mount on the insert body Repeat point 4 and 5 9 Dismount the insert head Take the two heaters and check their resistance you should read around 91 Q on the smaller inner heater and 65 on the larger outer heater 10 Mount the insert head on the insert body and repeat point 4 and 5 Attach bolts to the long insert head screws to secure it and connect one end of the heater current divider to the inner heater Screw the outer heater o
143. rich part of the Mn Fe P Ge system s phase diagram appears to have been largely disregarded as all recent publication only focus on the Mn rich samples as shown in Figure 7 1 Mn Fe P Ge 2 x x 1 y y 0 16 0 18 0 20 0 22 0 24 0 26 0 28 0 30 E s 7 c 0 75 0 80 0 85 Fe content x 0 95 oa pop P p ER 4 iam 0 16 0 18 0 20 0 22 0 24 0 26 0 28 0 30 Ge content y Ref 1 Ref 7 Ref 3 4 v Ref 8 e Ref 5 A Ref 9 10 m Ref 6 Figure 7 1 Studied compositions of recently published papers on the Mn Fe P Ge system For this purpose various Fe rich samples belonging to the Mn Fe 2 P Ge system and the closely related Mn Fe 95 P Ge system have been produced to evaluate their transition temperatures and overall magnetic properties 88 7 2 Material overview Both the Mn Fe P Ge and the Mn Fe 9s P Ge systems crystallize in the Fe P type hexagonal structure P6 2m space group 1 7 The Fe and Mn transition metal atoms occupy the 3f site at the tetrahedral x 0 0 position and the 3g site at the pyramidal x2 0 1 2 position The non metal P and Ge atoms can both occupy the 1b site at the 0 O 1 2 position and in the 2c site at the 1 3 2 3 0 position Considering the parent compound Fe P when Mn atoms are added to replace Fe they preferably occupy the 3g sites coplanar with the P and Ge atoms at the 1b site while the Fe atoms preferably occupy the 3f
144. rk Ref 12 Ref 14 Ref 19 Ref 21 Ferromagnetic 0 0 04 02 03 04 0 5 06 07 08 09 1 0 Fe Si Fe P Figure 5 7 Magnetostructural phase diagram of the Mn Fe Si P system revealing the compositional areas of all the different crystal structures described in the current chapter and their magnetic behaviors at room temperature for an annealing temperature of 950 C The color code refers to transition temperatures Ty for the antiferromagnetic samples and Tc for the ferromagnetic samples In the canted ferromagnetic phase area referred to in this diagram as the Antiferromagnetic canted ferromagnetic area only the values of Tc have been inserted in the diagram seeing as Ty here is relatively constant It should be noted that all border lines presented are only estimates made through the analysis of phase fractions on each sample 5 5 Overall diagram description The MnsSi compound orders in the cubic Fm3m structure and is an antiferromagnet below a Ty of about 25 K and paramagnetic above this temperature This behavior is maintained for P substitutions up to 0 2 where we find the tetragonal 14 structure This structure is then observed until the full substitution of Si by P in the Mn P compound with an increase in Ty up to 115 K for Mn3P With increasing Fe substitution the cubic structure maintains its border along the Mn Fe Sio Po line and displays a slow increase in Ty until a Fe content of 1 2 The tetragonal
145. rrectly then all should boil down to adequately operating the Calorimetry vi interface software A 5 1 Software handling 146 Open the Calorimetry vi Before turning any of the blocks on go through the following steps 1 Check the save time in the saving block This should read 10 seconds 2 Check the saving time on the Keithley 2400 block the Delay box this should at least read 1000 ms or 1 second 3 Check the saving rate in the Keithley 2000 block the Delay box this should at least read 50 ms This value is extremely relevant 4 Check the Read Setpoint switch on the Lakeshore 331 Temperature Controller block this should be turned up 5 Check the current work folder for any previous measurement files these will be named meas tdms and meas tdms index Since the data recording is done on a substitution basis it is essential that these be removed Ignoring this will result in files with mixed data from various different measurements Should all of the above have been observed and given the semi automated nature of most of the measurement blocks you should be ready to start a measurement Turn the Labview program on and in a rapid sequence turn all the measurements blocks on The order and speed in which you do this is not necessarily fundamental as during the subsequent data analysis all the data from the various blocks will be align among itself in order of Time This Time it is important to note i
146. rs as we know today first came about in the early 20th century These where mainly operated with the vapor compression method using steam engines with open drive compressors operating with dangerous and environmentally unfriendly refrigerants being anyway too large dangerous and expensive for a wide range distribution 1 Later in 1930 this system evolved and the use of CFCs Chlorofluorocarbons in refrigeration rapidly dominated the market Still latter research revealed that the use of uncontrolled CFCs was significantly hazardous to the stratospheric ozone layer and due to the Montreal protocol the use of these gasses was substituted by that of HFCs Hydrochlorofluorocarbons Although these do not damage the ozone layer they contribute to the rise of the earth s average temperature and to the greenhouse effect Both of these were later targeted by the Kyoto protocol and governments around the world became committed to reducing the use of these gases 2 opening the field for a new environmentally friendly technology to come along Considering that electricity consumption represents one fifth of the total energy used in Europe 3 and that modern day refrigerators work well below the optimal Carnot cycle 4 an improvement on this technology has to reflect not only environmental friendliness but also energetic efficiency Magnetic refrigeration based on the magnetocaloric effect MCE provides a new and viable solution for the creation
147. s set by an autonomous clock and as such it will always be the same and coherent for all blocks None the less the order given in Figure A 26 has been determined as the one which is less likely to give you problems related to unfinished modules processes in the long run 147 gt 1 13pt Application Font Ille Gil mera jo E o S T 6 Figure A 26 Suggested sequence for turning the different measurement blocks on Once you have done this the interface software should display an arrangement similar to Figure A 27 on Bl 555520000 Fans 5435922624 Jg EE Figure A 27 A functioning and measuring software interface Carefully check each block and their corresponding modules for malfunctions with the following steps 148 1 All wellness lights should be green 2 The overall save light should flash every 10 seconds if this is the defined time 3 The save light in the Lakeshore 331 Temperature Controller block should be flashing at a high rate 4 There should be values in the four boxes of the Lakeshore 331 Temperature Controller block and these should correspond to the values being presented in the 331 front panel as exemplified by Figure A 28 Figure A 28 Comparison between the Lakeshore 331 Temperature Controller block and front panel during a working measurement 5 The Keithley 2000 should be
148. s of a small 25 mm wide cylinder protected by three cylindrical shields of which the first of these shield is equipped with a heater A scheme of the insert head is shown in Figure 4 2 35 Figure 4 2 a Insert head scheme b Experimental sample space Legend A Body of the insert B Connector enabling the head to be detached from the insert body and easily handeled C Wiring connection to the heating cylinder D Vacuum cylinder E Calorimetry chip socket F Xensor microcalorimetry chips see below G Soldering platform functioning as a thermal anchor for all the wiring inside the insert head H Wiring between the chip sockets and the soldering platform I Wiring from the soldering platform to the body of the insert J Wiring from the heating cylinder to the body of the insert K Heating cylinder L Carbon glass temperature dependent resistor M Wiring from the resistor to the soldering platform 4 2 2 Temperature control All aspects of temperature control in the current setup are done by a Lakeshore 331 Temperature Controller This instrument measures the resistance of the carbon glass temperature dependent resistor placed next to the calorimetry chips and in thermal contact with the heating shield using a 4 point measurement method This resistance value is then converted into a temperature through a calibration table previously measured and displayed in figure 4 3 The determination
149. s shown great promise having a highly tunable transition temperature low thermal hysteresis and high magnetization 1 This system however has the drawback of being considerably more expensive due to the use of germanium which limits its potential use in commercial magnetic refrigeration devices The results presented by Trung et al 1 and more recently by Dung et al on the Mn Fe 9s P S1 system 2 do reveal that increasing the Fe composition in these systems triggers an increase in their transition temperature which may then be compensated by a decrease in Ge or Si content Obviously among these two systems Mn Fe os P Si stands out as the cheapest and safest possessing remarkable magnetic and magnetocaloric properties Unfortunately the dramatic change in the a and c parameters observed in its crystal lattice during its first order magneto elastic transition implies that these materials may display a potentially crippling brittleness when they are thermally or magnetically cycled 2 which is an added difficulty for their commercial application This has lead to the motivation to investigate Fe rich Mn Fe P Ge samples in more detail as the characteristics of this system s magneto elastic transition suggest a more stable mechanical behavior and the above mentioned results seem to point to the fact that Ge content can be significantly reduced if Fe is increased possibly making this system commercially viable The Fe
150. s vale agradecer quelas pessoas que n o saber o ler nada do que l vai em cima o que n o grande perda diga se de passagem Ora acho que o obvio ser come ar pelos meus santos pais eu sei a todos os momentas da minha vida que o vosso amor e apoio s o incondicionais e se alguma vez pareci emocionalmente distante foi precisamente por ter esse facto como certo 175 Jo o na verdade n o me ajudaste neste trabalho mas ajudaste de maneiras muito significantes em muitos outros que decidi arcar contra os meus pr prios concelhos E isto ao ponto de eu dizer p j chega o que no fundo apenas quer dizer que foste sempre prest vel de mais No fundo fora isto n o mantive amizades com estudantes Portugueses por estas bandas Em tempos tentei mas r pido me apercebi que a qualidade de gente da minha terra que por esta terra anda tinha orienta es diferentes das minhas Da vossa boca apenas ouvi referencias a pulhice sua canalhada pueril da puta que vos pariu E assim sendo apenas me posso referir queles outros Portugueses que n o caiem nesta categoria ou seja aqueles que n o est o neste pa s ou aqueles que est o j h anos demais para o seu pr prio bem mas n o todos que um dia vi um velho no aeroporto que merecia uma chapada na boca F bio p ainda que do outro lado do Mar do Norte o contacto que mantivemos ao longo destes anos foi para mim uma refer ncia essencial Nestas outras lides n o teria cheg
151. s were reduced to a fine powder and firstly mixed with varnish to solidify the powder with random grain orientation random powder For comparison the same process was followed but the varnish solidified while the powder was under a magnetic field meaning that the grains were able to align with this field field oriented powder Figure 7 11 displays the results from these measurements 99 ee aaa ee ar en 0 8 a een a 5K T 430 K T 515 K Mn Fe ssPo Ge a Mn Fe aso 85885 15 s Random powder s Random powder Field onented powder 4 Field oriented powder 0 0 95 10 15 20 25 30 35 40 4 5 50 5 5 0 0 0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 Magnetic Field T Magnetic Field T C 10 AAA pai and dae e E gd 0 84 ae A a 0 6 5K 3 0 44 Ee me T 7455K 0 24 Fe P Ge s Random powder 0 0 A hield oriented powder 0 0 05 10 15 20 25 30 35 40 45 50 5 5 Magnetic Field T Figure 7 11 a Magnetization versus applied magnetic field at 5 K for the Mn Fe 4 Po4Geo sample As can be observed there is a clear difference in magnetic behavior between the field aligned and the randomly alighted sample indicating the presence of strong magnetic anisotropy b Magnetization versus applied magnetic field at 5 K for the Mn Fe s Po5 Geo 5 sample demonstrating a similar but not as pronounced m
152. specific heat measurements as they then represent measurements with an applied field along the hard axis figures 8 2 and 8 3 This difference in orientation between the applied magnetic field and the easy axis of the samples can be easily explained by the particularities of sample mounting in 111 both equipments used The measurements of Figure 8 4 a were taken in a MPMS XL SQUID on which samples are mounted vertically with the applied magnetic field along the c axis of a single crystal while the XEN 39328 microcalorimetry chips used for Figures 8 2 and 8 3 mount samples horizontally which when inserted into the AMI Cryostat of the experimental setup places the easy axis perpendicular to the applied magnetic field as demonstrated in Figure 8 5 b easy B direction Figure 8 5 a Vertical sample mounting on a MPMS XL magnetometer placing the easy axis of a single crystal along the applied magnetic field direction image taken from 22 b Horizontal sample mounting on the XEN 39328 microcalorimetry chips placing the easy axis of a single crystal perpendicular to the applied magnetic field image taken from 23 It should further be noted that our equipment and sample handling protocol is not prepared to make distinctions between easy and hard directions and such a measurement was unintended 8 5 Conclusions In order to further investigate and understand the properties of Fe P the parent compound of muc
153. sult may be a direct consequence of the different sample preparation procedures used in the current chapter and those used by the mentioned authors namely arc melting annealing for 24 hours at 800 C and quenching into cold water 51 On the Mn rich side the 1 4 tetragonal phase was found to exist up to a composition of MnsSiosPo On the Fe rich side this structure was found to exist up to the Si substitution of Fe3Si04Po 6 The diffraction data from samples displaying the orthorhombic phase was found to be consistent with the Pmmm space group Figure 5 3a This structure was also found for the MnFe Sio sPo s composition A fourth crystal structure consistent with the hexagonal P6 mmm space group was found for values in the range of 0 2 lt x lt 1 95 and 0 2 lt y lt 0 9 for Mn Fe Si P as shown in Figure 5 3b It is possible that this structure is the same as the one observed by Yoon et al 12 as a secondary phase in cubic samples with Mn content above 1 8 but as these authors did not provide any additional information regarding this secondary phase we cannot confirm this hypothesis The detection of this structure in this system can be said to be a complete novelty and has never been reported in literature The lattice constant a for the cubic phase has been found to decrease with the Fe content ranging from 5 72 A for MnsSi to 5 65 A for FesSi in excellent accordance with the values found by Yoon et al 12 However the s
154. switching between channels 3 and 4 in a rate of about 30 seconds and you should be seeing a very clear voltage response to the wave being generated by the Keithley 2400 see below on both the Keithley 2000 block and the Keithley 2000 front panel as exemplified by Figure A 29 149 2000 MUL TIME Figure A 29 a Comparison between the Keithley 2000 software block and front panel while on channel 4 and under the influence of a wave from the Keithley 2400 b Comparison between the Keithley 2000 software block and front panel while on channel 3 and not under the influence of a wave from the Keithley 2400 6 The Keithley 2400 front panel should be switching between 0 mA and a concrete value this will depend on your chip resistance as exemplified by Figure A 30 150 Figure A 30 a Comparison between the Keithley 2400 software block and front panel while not producing a square wave b Comparison between the Keithley 2400 software block and front panel while producing a square wave 7 There should be new meas tdms and meas tdms index files in the work folder Should all of these be observed the measurement is working on all accounts 151 A 5 2 Temperature sweeps The above instructions hold true for measuring in the abstract sense but a great deal more needs to be added to them in order to perform a useful calorimetric measurement In order to perform a temperature sweep you will have to work in part with the in
155. t being performed there might be the necessity to change this range in mid measurement There are two ways of handling this problem the manual solution and the Zone Setting solution The manual solution is straight forward arrange your measurements in such a way as it will be possible to do the heater range change manually Remembering that 100 of any heater range is 10 of the next one simply do this change when you know that the heater setting you are defining is able to pick up from were the previous one left off For example when you reach 10 of the Low setting you should be able to switch to the Medium which will then translate itself as 1 of this setting being the same also valid for the transition between Medium and High Since you are doing a measurement this change needs to be done in the software interface The zone setting solution is a much more convenient way of solving this issue even if it requires some additional work As this feature of the Lakeshore 331 Temperature Controller is not specific to the current microcalorimetry instrument you should refer to this module s User s Manual for more details A 5 3 Magnetic fields Currently the magnetic field control modules are not in any way related to all the rest of the control and measurement modules and software discussed so far This means that there is no possible remote control of this equipment and that the calorimetry measuring and field control systems shoul
156. t het bestaan van een Fe Co 3 S P systeem zien welke een afstembare Tc heeft met S gehalte evenals een nieuwe overgang van onbekende aard bij 370 K Hoofdstuk 7 exploreert de Fe rijke kant van het Mn Fe P Ge magnetocalorische systeem De overgangs temperatuur van dit systeem is buitengewoon eenvoudig af te stemmen door een zorgvuldige manipulatie van het Fe en Ge gehalte en de stochiometrische verhoudingen wat leidt tot de re le mogelijkheid de prijs van de benodigde Ge te verlagen en de verbinding economisch haalbaar te maken voor praktische magnetocalorische toepassingen Terwijl het Mn Fe P Ge systeem lijdt onder een significant verlies aan scherpte van de overgang bij verlagen van het Ge gehalte werd dat niet vastgesteld bij het Mn Fe os P Ge systeem en bij het laagst mogelijke Ge gehalte die samenvalt met de laagste mogelijke financi le kosten was de Tc af te stemmen onder een temperatuur van 225 K met de magnetische entropie verandering schommelend tussen 3 en 3 5 J Kg K bij een veld van 2 Tesla Nieuwe en onverwachte magnetische eigenschappen die in dit systeem zijn waargenomen suggereren een opwindend potentieel voor permanente magneet toepassingen in een bepaalde Mn en Ge arme concentraties Hoofdstuk 8 geeft de microcalorimetrie resultaten aan zuiver Fe P gemeten met de experimentele installatie zoals beschreven in hoofdstuk 4 Deze laten het geleidelijke verlies van het le orde karakter van de overgang van deze v
157. terface software and with the actual physical Lakeshore 331 Temperature Controller as the interface does not possess all the necessary tools to do this remotely To set a temperature ramp follow the below instructions 1 Be sure that the Remote light of the Lakeshore 331 Temperature Controler is turned off if not turn it off by pressing the Remote Local button 2 Be sure that the Ramp light is turned off If it is not press the Control Setup button followed by Enter until you read Setpoint Ramp On Press one of the arrow buttons so this reads Setpoint Ramp Off and press Enter The two points above are preliminaries in order to simplify the next steps 3 Press the Setpoint button and using the number pad insert the initial temperature for your ramp this should be the temperature the sample space is in 4 Repeat the steps in point 2 so as to turn the Setpoint Ramp Off into Setpoint Ramp On When you do this you will read Ramp Rate on the Lakeshore 331 Temperature Controller display Using the numerical pad enter your ramp rate and press Enter The ramp light should now turn back on 5 Go to the interface software and be sure that the Setpoint Set in the Lakeshore 331 Temperature Controller block is the same as your initial ramp setpoint which you should be reading in the front Lakeshore 331 Temperature Controller panel 6 Start the measurement as described in section A 5
158. the preparation of lab scale intermetallic compounds with a maximal mass of 15 grams The electrical current in this setup is tunable between 5 and 300 A 2 3 All samples produced with this equipment were melted on a water cooled copper crucible inside the arc melting furnace under a 500 mbar atmosphere of purified argon The as melted ingots were turned over and re melted again between 3 and 5 times to attain good homogeneity before annealing 26 1 Stainless steel bellows 2 Lift 3 Plexyglass tube for safety 4 Pyrex tube 5 Water cooled tungsten electrode 6 Water cooled copper crucible 7 Manipulation ring 6 Pressure indicator 9 Vacuum connection for casting 10 Turbo pump 11 Pneumatic valve 12 Electronic equipment 13 Water leak detector 14 Three way valve 15 Fore pump Figure 3 2 Schematic representation of the home made arc melting furnace used in sample production 2 3 4 X ray diffraction The crystal structure and sample homogeneity were analyzed with an X Pert PRO X ray diffractometer from PANalytical using Cu K radiation schematically represented in Figure 3 3 both in ambient and non ambient conditions only used in Chapter 8 The resulting diffraction patterns were analyzed using the X Pert HighScore and FullProf s software implementation of the Rietveld refinement method 4 27 Detector slit X ray tube Receiving slit Soller slit Anti scatte
159. the respective legend given in Table A 7 132 Figure A 21 Saving interface software block Table A 7 Legend for Figure A 21 Global save periodicity This box is directly related to an unexplained inbuilt error of LabView This translates itself as the fact that a measurement file needs to be closed and opened occasionally so as there may be a release of computer memory If this does not happen given the long nature of the measurements performed in this setup the computer will run out of memory crash and lose the measurement 2 Global saving light This light should flash according to the global save periodicity 3 Error light Global stop button This shuts down the whole interface software A 3 2 Data visualization and extraction software This program does not need to be installed in the same computer onto which the three Keithley modules and the Lakeshore 331 Temperature Controller are connected via a GPIB bus It is designated as Calorimetry read vi and upon opening it one should be confronted with a panel such as Figure A 22 with the respective legend given in Table A 8 133 el e n Besier L EO eit Input File Path Save File amp D SharedData calorimeter Peter Jose cooling_test_meas tdms 1 D sharedData calorimeter Peter Jose test measurments test tdms 2 4 Channels x Channels y 3 EIS Crem Coa 0 01 0 00 0 01 0 02 0 03 0
160. thermopiles 10 the temperature dependent resistance of the chip heaters and other bothersome calibrations often required for temperature scanning experiments To insure good thermal contact between the heaters on the chips their sensors and the sample Apiezon N and H greases are used for low and high temperature measurements respectively This further prevents sample motion during chip handling and application of the external magnetic field The specific heat of Apiezon N grease has been studied extensively 18 21 and can easily be taken into consideration during the data analysis This issue is further reduced due to the double chip setup when the amount of grease in both chips is comparable 4 2 4 External measurement and oower supply equipment The power supplied to the chips comes from a Keithley 2400 Source Meter To measure and register the voltage output of the calorimetry chips thermopiles a Keithley 2000 is used This is equipped with a scanner card enabling us to switch between different channels and measure different signals thus simultaneously measuring both the reference and the sample chip The information supplied by the previous two modules and the Lakeshore temperature controller is then fed into a PC via a GPIB bus where the data is processed and analyzed via LabVIEW and MATLAB routines 4 3 Measurement cycle The specific heat in this setup can be determined by the so called relaxation method In this method i
161. tion and decrease with Fe while the c parameter decreases with Ge and increases with Fe The Tc of these systems was found to be quite easy to manipulate and tune with changes in both Fe and Ge content For the Mn Fe P Ge system the diagram in Figure 7 4 was assembled The results displayed in Figure 7 4 show that in this compositional range Tc is easily tunable but does not change linearly with Fe or Ge concentration in contrast to what had previously been reported by Trung et al 1 and Br ck et al 14 for Mn rich Mn Fe P Ge compounds Novel in this system is the detection of an anomalous magnetic behavior at low Ge concentrations not consistent with regular full Ferromagnetic behavior If pushed to even lower Ge contents this turns into full Antiferromagnetic behavior limiting the lowest usable Ge content in this system This novel magnetic region seems to be influenced by the Fe content as increasing Fe pushes the occurrence of this behavior further into low Ge concentrations eventually vanishing from our samples A similar behavior to this seems to have been also observed by Tegus et al 15 16 both in the Mn Fe P As Ge and Mn Fe P As systems for samples with low As and Ge content The exact nature of this behavior has not yet been established but given the contraction of the lattice parameter a and expansion of c also observed in Mn Fe P As 16 this phenomenon should associated with intra layer interactions b
162. tions Bodmin MPG Books Ltd 2003 K A Gschneidner Jr V K Pecharsky and A O Tsokol Rep Prog Phys 68 2005 1479 Anders Smith Christian R H Bahl Rasmus Bjgrk Kurt Engelbrecht Kaspar K Nielsen and Nini Pryds Adv Energy Mater 2 2012 1288 Jordi Marcos F lix Casanova Xavier Batlle Am lcar Labarta Antoni Planes and Llu s Mafiosa Rev Sci Instrum 74 2003 4768 V V Korolev A S Romanov and I M Aref ev Russ J Phys Ch 80 2006 310 Michaela Kuepferling Carlo Paolo Sasso Vittorio Basso and Laura Giudici IEEE T Magn 43 2007 2764 G Porcari F Cugini S Fabbrici C Pernechele F Albertini M Buzzi M Mangia and M Solzi Phys Rev B 86 2012 104432 Giacomo Porcari Magnetocaloric effect across first order transformations of energy conversion materials Universit degli studi di Parma 2013 PhD thesis Chapter 3 S Jeppesen S Linderoth N Pryds L Theil Kuhn and J Buch Jensen Rev Sci Instrum 79 2008 083901 K Morrison Y Miyoshi J D Moore A Barcza K G Sandeman A D Caplin and L F Cohen Phys Rev B 78 2008 134418 11 12 13 14 15 16 17 18 19 20 21 22 23 K Morrison J D Moore K G Sandeman A D Caplin and L F Cohen Phys Rev B 79 2009 134408 A A Minakov S A Adamovsky C Schick Thermochim Acta 432 2005 177 A A Minakov S B Roy Y V Bugoslavsky and L F Cohen Rev Sci Instrum 7
163. tlab and set your work folder to the folder where you have your data files making sure the two grabtau m and capacity temperature matcher m routines are also placed in this folder 8 Run the grabtau m 9 This should open two image windows Image 1 with an assortment of graphs and Image 2 with a small selection of your chip thermopile responses not important Image 1 should be the one you pay more attention to and it consists of a number of graphs detailed in Figure A 37 with the respective legend given in Table A 10 161 Measurementchannel3 csv x e ze T T T T T T T T T S 1 S 4 i I TUM AAA ERN RIMINI lil IN Ul IN HH nH DRIN n DINI BER 0 02 04 0 6 0 8 1 42 14 1 6 18 2 en x10 Parameterization Initial voltage Parameterization Decay time e m T E o E 3 E E 8 3 a 5 2 8 0 02 04 06 08 1 12 044 t5 AB 2 4 x10 Approximate heat capacity x 10 25 T T T T T T T T T Worst fit K A 5 2 s 2L w Measurements qe verae re Fit S LS SIS F Ne cpl Na re E Se ine aa aaa 2 H SN 0 1 fi 1 1 L 1 L L 1 gt A 0 02 04 06 08 1 T5 4 t6 08 2 5 E z Sa Heat capacity ratio x10 3 L 6 E 5 BB E 10t Negra AMA ak y S E A d Eq e 12 is amp 3 6 I 2 44 1 1 1 1 1 1 L 1 1 1 1 N 1 1 1 1 1 88 90 92 94 96 98 100 102 104 0 02 04 06 08 1 2 14 16 48 2 4 x 10 Figure A 37
164. to 450 K 14 and still maintain its hexagonal structure This family of compounds is the favored one by the Delft magnetocaloric group and either directly Chapters 7 and 8 or indirectly Chapters 5 and 6 is the driving motivation behind most of the research presented in this thesis 1 4 Thesis outline This thesis is partially focused on research performed in the scope of the optimization of Fe P type compounds Apart from this there is also a great deal of focus on the assembly and use of a home made experimental setup for in field microcalorimetry measurements More specifically Chapter 2 covers the necessary theoretical concepts for the understanding of the MCE as defined above It also gives an overview of the magnetic anisotropy phenomenon which is relevant for the understanding of certain results presented in the following chapters Chapter 3 covers the main experimental procedures and equipment used during sample preparation and characterization Chapter 4 focuses particularly on the already mentioned home made in field microcalorimetry setup assembled during the course of this study Chapter 5 describes our findings in the Mn Fe 3 Si P system which is related to the Mn Fe Si P systems As a result of this study resorting to the extensive experimental characterization this system went through a magnetostructural map was assembled which includes the discovery of a novel hexagonal phase Chapter 6 continues the exploratio
165. to calculate the actual adiabatic temperature change of a material as well as providing reliable and precise information on any phase transition that may be influenced by the application of a magnetic field In Chapter 5 the Mn Fe Si P system is fully explored with X ray diffraction DSC and magnetization measurements This system has been determined to be inadequate for magnetocaloric applications but the research performed allowed for the assembly of a magnoestructural map of this system Besides the already known cubic phase for Mn Fe Si system and the tetragonal and orthorhombic phases for the Mn Fe P system a novel hexagonal phase has been observed for Mn Fe Si P within the approximate range of 0 2 lt x lt 2 0 and 0 2 y 0 9 Magnetization measurements both confirm and further detail the already known properties of the Mn Fe Si and Mn Fe P systems Given the interesting character of an antiferro forremagentic transition present in Si rich samples with Fe content ranging from 1 2 to 2 3 a study on the effect of interstitial Boron was also performed on this system This resulted in the shift of this transition from 50 5 to 57 K for the B content of 0 05 with all higher concentration resulting in the occurrence of secondary phases Chapter 6 is an attempt at studying the novel hexagonal phase discovered in Chapter 5 by the introduction of Co into Mn Fe Si P system creating in this way the Mn Co Si P and Fe Co 3 Si
166. ubstitution of P seems to promote an increases in the lattice parameter a as both for FesSio sPo2 and FesSio sPo s a value for a of about 5 66 was found The a and c lattice constants in the Mn rich tetragonal phase is also in good accordance with the literature 20 21 as both decrease with increasing Fe content Si substitution increases both a and c as a 9 179 and c 4 568 A for Mn P while a 9 183 A and c 4 607 A for Mn3Sio 2Po s On the Fe rich tetragonal phase values a 9 104 and c 4 4631 were found for the FesP compound in accordance with literature 19 21 We were unable to establish whether the evolution of the lattice parameter with Mn substitution in this compositional area The presence of an orthorhombic secondary phase in our samples prevents us from achieving a good fit of our X ray diffraction patterns However it has been determined that Si substitution strongly increases both lattice parameters The composition of FesSio2Po g shows a lattice paramenters of an a 9 1 168 and c 4 4782 The lattice parameters of the orthorhombic phase were found to be slightly smaller than those reported by Goto et al 21 with a 8 9456 b 8 0079 and c 4 3368 A for the Mn sFe sP compound Due to the presence of secondary phases 52 belonging to both the Mn and the Fe rich tetragonal phase and the hexagonal phase we were able to make a good fitting of our other orthorhombic samples The hexagonal phase for Mn
167. ular silicon 99 red phosphorous powder and 99 9 manganese chips reduced at 600 C under a hydrogen atmosphere in order to remove oxides The samples where ball milled as described in Chapter 3 then compacted into 10 mm pellets with a pressure of 150 kgf cm and sealed into quartz tubes with an atmosphere of 200 mbar of argon Finally these were annealed in a vertical resistive furnace for 100 hours at 950 C and quenched in water at room temperature This procedure was selected to be appropriate for all possible composition variations of the Mn Fe Si P system The use of an arc melting furnace a common procedure for the production of samples belonging to the Mn Fe Si system 10 12 22 was found to be unsuitable for the production of samples containing Phosphorous Magnetic and DSC measurements where performed on all samples and homogeneity and crystal structure was checked using X ray diffraction as described in Chapter 2 5 4 Results and discussion 5 4 1 Structural results All the structural information regarding the Mn Fe Si and Mn Fe P systems mentioned in section 5 2 was confirmed The Fm3m cubic structure of the Mn3Si and Fe3Si was found to exist up to the Phosphorous substitution of MnsSio sPo2 on the Mn rich side of the diagram and up to Fe3Sio 4Po on the Fe rich side The secondary tetragonal or hexagonal phases reported by Yoon et al 12 in cubic samples with a Mn content above 1 8 were not observed Such a re
168. v S B Roy Y V Bugoslavsky and L F Cohen Rev Sci Instrum 76 2005 043906 Addendum Instruction Manual for the use and handling of the microcalorimetry setup described in Chapter 4 A 1 Hardware overview The total sum of all equipment involved in the current microcalorimetry instrument can firstly be divided into two categories 1 The setup This consists of the AMI cryostat equipped with a 9 Tesla 2 inch bore superconducting magnet and the home made insert fitted for microcalorimetry measurements represented in Figure A 1 Figure A 1 a Schematic representation of the AMI cryostat b Schematic representation of the home made insert 2 The outside of the setup This consists of the several measurement and control modules namely Lakeshore 331 Temperature Controller both used to measure the sample space temperature and supply power to the sample space heater s represented in Figure A 2 116 Como low apt vp Reman wo ou SGG UN 1 Es Figure A 2 Lakeshore 331 Temperature Controller Keithley 2400 Source Meter used to supply and measure voltage to the Xensor microcalorimetry chips heaters used in the setup represented in Figure A 3 In the current case relaxation measurement it supplies a square wave ma ocx re CT n Cora me dei gt L auto DEE n RANGE o cm sreeo on en menu ext enter Figure A 3 Keithley 2400 Source Meter Keithely 2000
169. whole microcalorimetry instrument It is here that all measurements and data acquisition happens For further clarification the insert head can be divided into the following subsections 1 The sample space containing the microcalorimetry chip socket platform the calorimetry chips and a temperature dependent carbon glass resistor 120 2 The insert head extension which is merely a carefully measured elongated piece of metal meant to center the sample space in the magnetic field produced by the 9 Tesla superconducting magnet 3 The insert head plug where all the wiring from the sockets chips and the resistor connect to These divisions are demonstrated in Figure A 10 Insert Body Temperature 7 Dependent Resistor Calorimetery Chips Chip Platform Figure A 10 a Insert subdivisions b insert head subdivision c sample space layout The platform on which the two removable Xensor microcalorimetry chips are meant to fit into is wired according to Figure A 11 with the respective legend given in Table A 1 121 Keithley 2400 Source Meter Figure A 11 Chip platform wiring The definition of Right and Left presented in this scheme will from this point on be used to distinguish both the chip sockets and the chips installed in the setup Note that this wiring was planed in accordance with the layout of Xensor microcalorimetry chips mounted on TO 5 frames Table A 1 Legend for Figure
170. wo heaters and re measure their resistance if their values are good then you have a wiring problem somewhere on the outside of the insert head Figure A 25 illustrates the position of the above mentioned buttons on the Lakeshore 331 Temperature Controller LakeShore 331 Temperature Controller ContolA Tune Remate ONDNDDDDDDNNONDEOEOEN 2 2 ODI EDDIE ee Control Zone Input Display mote Setu Setin Setu Fomat Aam Local 2 La Ce d 2 Interface PID Curve Analog MHP Ent mous Output 51 Lel CH Figure A 25 Location of all the buttons mentioned above in the Lakeshore 331 Temperature Controller front pannel 12 Take the large outer shield that covers the insert head and thoroughly clean its top part Using Apeazon N or H depending on the nature of your measurement coat the outer indented ring on the top of this shield Cut a piece of Indium wire of 18 5 cm and carefully fit it to the outer ring This is meant to function as an O ring for the vacuum created in the sample space during a measurement 13 Carefully attach the outer shield to the insert body trying your best to make sure the indium gets pressed homogeneously all around the shield by slowly tightening its screws in opposing pairs 145 14 Repeat point 4 5 and 11 15 Is there either liquid Nitrogen or Helium inside the cryostat No Gently lower the insert into the cryostat taking special attention to tilt it as little as possible As soon as it is
171. y mentioned second 74 phase which displays a much lower magnetization Tc was found to be beyond the range of both our magnetometers and our DSC measurement system but as we increased the P content an additional weak magnetic transition became visible as a result of the formed second 2 1 phase which exhibits a low and quite tunable Tc 19 21 6 3 2 P rich samples Given the high Tc of the Co and Si rich cubic phase samples the Fe and P rich side with a tetragonal structure is most likely to provide candidates for a successful magnetocaloric approach Coming from the pure Fe P compound with a Tc of 686 K 8 10 we explored the Co dependence of Tc in order to tune down this transition to a more accessible temperature In contrast to what has been reported by Goto et al 8 and Liu et al 9 we were able to produce single phase samples with a Co content above 30 substitution The results are quite close to the phase diagram presented by Raghavan 22 shown in Figure 6 2 As such we were able to lower Tc below the reported 640 K 8 9 to around 600 K as shown in Figure 6 3a To complement the results by Liu 9 we monitored the evolution of the lattice parameters a and c with increasing Co as shown in Figure 6 3b Further attempts to manipulate Tc by substitution of P by Si presented multiple phases These results and those of the Si rich side of the system may be taken as solid evidence that the quaternary Fe Co Si P does not
172. y of Aveiro where he took his bachelor degree in theoretical physics finishing in 2007 under the supervision of Prof Vitor Amaral and Dr Mario Reis In that same year he followed a Master degree in this same university also in theoretical physics finishing in 2008 having his final project on the magnetocaloric effect once again supervised by Prof Vitor Amaral and Dr Mario Reis In 2009 he was accepted as a PhD student in the Delft University of Technology under the supervision of Prof Ekkes Br ck with his assigned project being a mix of magnetocaloric material exploration and the assembly of an experimental setup and whose result are presented in the current thesis Parallel to his daytime academic career he regularly publishes essays and papers in literature and philosophical periodical journal having also broken into the worlds of occult publishing His particular interest in Portuguese and Ibero Afro American forms of spirituality and transcendence has also led him to further lecture and publish in academic anthropological seminars 184 Epilogue Inquieta o A contas com o bem que tu me fazes A contas com o mal por que passei Com tantas guerras que travei J n o sei fazer as pazes S o flores aos milh es entre ru nas Meu peito feito campo de batalha Cada alvorada que me ensinas Oiro em p que o vento espalha C dentro inquieta o inquieta o s inquieta o inquieta o Porqu n o sei Porqu
173. yostat and insert As a base for this setup we resorted to a commercial cryostat from American Magnetics Inc AMD equipped with a 9 Tesla 2 inch bore superconducting magnet Solenoid with its own power supply and magnetic field controller This cryostat has a 36 1 LHe reservoir in direct contact with the superconducting magnet to keep it at a constant temperature of 4 2 K A separate LN reservoir with a capacity for 36 1 is also present to reduce helium evaporation The Variable Temperature Insert VTI fitted for transport measurements was removed to mount another insert with the capacity to perform specific heat 34 measurements shown in Figure 4 1a This step made it impossible to use the original built in temperature control system of this cryostat A schematic diagram of the cryostat is shown in Figure 4 1b Figure 4 1 a Comparison of the original transport measurements VTI insert and its respective casing left and middle respectively with the newly fashioned home made microcalorimetry insert right b Schematic diagram of the AMI cryostat Legend A Liquid Nitrogen reservoir B Liquid Helium reservoir C Insert D Magnetic field center E Superconducting Magnet F Wiring conecting the insert to the rest of the measurement equipment vacuum pump tube The evacuated sample space in the tip of the newly constructed insert is in direct contact with the liquid Helium and consist

LEITAO_THESIS - Repository TU Delft

Contents

Download Pdf Manuals

Related Search

Related Contents