PROTON MAGNETSC RESONANCE SPECTRA. OF GASEOU‘S AND HOW?) HYDROCARW Thesis f’or Hm Degree of M. S IVIICHEGAN STATE UNWERSETY Norman R Lame 1958 THESIS —-—-- -‘ “a f) LIBRARY {‘1 Michigan State 37. University PROTON ”XGNETIC R£SON3NCE SPECTRX OE GXSECUS \ND LIQUID HYDROCXRBONS By Norman R. Laine P P3 III Fl 1 I) H 01 Submitted to the College of Science and irts of Michigan State University of igriculture and ipplied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1958 XBSTRACT The nuclear magnetic resonance Spectra of simple gas- eous and liquid hydrocarbons were observed under high reso- lution by use of the model V-4300—2 Varian Spectrometer sys- tem. A Special sample container was constructed of Teflon to contain the gaseous samples under pressures up to 40 atmos- pheres. Liquid samples were contained in sealed Pyrex tubes. The chemical Shifts (J) of the peaks in the Spectra of liquid prepane, prOpylene, prOpene-2-d, prOpyne, allene and the isomeric butenes were measured with reSpect to benzene and those of gaseous methane, ethane, ethylene, acetylene, prOpane, prOpylene and cyclOprOpane with reSpect to hydrogen gas mixed with the gas being studied. Where possible nuclear spin-Spin coupling constants (J) for the electron-coupled ‘ Spin-Spin interactions between protons in neighboring groups have been obtained also. The Spectra have been interpreted in terms of the chem— cal shifts and Spin-Spin coupling constants. In the case of prOpane the chemical shift and Spin-Spin coupling constant were found by the method of Anderson and McConnell (7); from the former a value of the electronegativity of the CH3 group has been derived. 11 ACKNOWLEDGEMENT The writer wishes to eXpress his sincere appreciation to Professor M. T. Rogers for his guidance and helpfulness and to Dr. P. T. Nara- simhan for his help and suggestions throughout the course of this work and also to the itomic Energy Commission for a grant supporting this work. -)(--).’--X--3<*-31-*-35** ***%**** ewes-recur: ‘1 \I \r '3’ w.» 7 iii VIII I}; TABLE OF CONTENTS PAGE H I INTRODUCTION...................................... II THEORETICAL.......................................I 3 III HISTORICAL........................................ 10 IV EXPERIMENTlL...................................... 16 v RESULTS........................................... 23 ‘VI CiLCULiTIONS...................................... 38 VII DISCUSSION........................................ 43 VIII SUMMARY........................................... 50 IX REFERENCES...000000000.000000000000000000000000.o. 52 iv TABLE II III IV VI VII VIII LIST OF TIBLES PiGE Proton Chemical Shifts of Substituted Hydro- carbonS........................................... 12 Proton Chemical Shifts of the Halomethanes........ 13 Proton Chemical Shifts and Electronegativities of Substituents in some Substituted Tthanes.....,.. 14 The Electronegativity of the Halogen Groups in the Ethyl Halides................................. 15 Boiling Points, Vapor Pressures and Purities of Samples Used................................... 21 Chemical Shifts of Gaseous Hydrocarbons........... 35 Chemical Shifts of Liquid Hydrocarbons............ 36 Relative Intensities and Separations of the Propane Peaks...OOOOOOOOOCOOOOOOOOOO0.00.00.00.00. 39 FIGURE 1. 2. 3. 4. 19. 2’0. Orientations and energy levels for I=%............. LIST OF FIGURES Simplified apparatus for basic NMR SXperiment..... Diagram or 8&8 Sample cell...OOOOOOOIOOOOOOOOOOOO. Pressure cap for gas sample cell.................. Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum of of of of of of of of of of of of of of of H2 and CH4 gas mixture................ H2 and CZH6 gas mixture............... 02H4'and CH4 gas mixture.............. H2 and 02H4 gas’mixture............... H2 and cyclOprOpane gas mixture....... prOpane gas........................... propylene gaS....................c;... liquid propane......................t. liquid'allene......................... liquid prOpylene..fl................... liquid propane-Q’dooooooooooooooooowoo liquid propyne...OOOOOOOOOOOOO0.0..O... liquid butene’looocooooormoooooooocoo. ligflid tmn8-2-bu12oene. o o o o o o o o o o Q 0 O o O o liQUid Gig-2-buteneoooooocooooo.0000.0 Plot of electronegativity of substituents versus ACH3 —ACH2 for substituted ethanes............... vi PAGE 18 19 26 26 20 V 27 27 28 29 29. 30 31 31 32 34 37 a ll! villi-I v.1lIvflLlI.|.'. 1i... Ia‘! E!— : , , .V §. at. 13.... ..ulAL|lJ.v 1.. [ INTRODUCTION I,INTRODUCTION DeSpite the large amount of work done on nuclear mag—' netic resonance, very little work has been done on the sim- ple gaseous hydrocarbons such as methane, ethane, ethylene, acetylene, prOpane, 233;. The purpose of this investigation was to obtain chemical shifts and nuclear Spin coupling constants for these Simple hydrocarbons where conditions for theoretical interpretation should be most favorable. Because many of the hydrocarbons observed are gases, a cell was constructed to hold gases under a pressure high enough to give a sufficient concentration of nuclei to pro— vide an observable Spectrum. Those gases with low vapor pressures were condensed into a glass sample tube and the Spectra of the liquids observed. THEORET ICAL II THEORETICAL (1-4) A given nucleus will absorb a radiofrequency signal when the signal frequency corrSSponds to an energy differ- ence between the two nuclear Spin energy levels for the nu— cleus,in an applied magnetic field. The absorption of this ‘signal causes a transition to take place between two of the energy levels. The resulting transition is a nuclear mag; netic absorption line. If the nucleus has no Spin there will be no absorption of the signal. Because the nucleus has both mass and charge there is an angular momentum and magnetic moment associated with this Spin. The magnitude of the nu- clear Spin angular momentum is JIT3:IY h/2F’ and its pro- Jection in the direction of an external field is Mh/EIT (I is the Spin of the nucleus and is the maximum allowed value of the magnetic quantum number M). The magnetic mo— ment components in the direction of an external field are ’5 =1Vp A ‘ SIIFA‘ 0 since, ’5 z p-EEE- _ Mhe ’uz “' 21! me A. is the nuclear magneton and is 5.0493 X 10'2'4 ergs/ gauss. gI is an irrational number and is 5.58490 for protons. I is the gyromagnetic ratio, iLgL, the ratio of the mag- netic moment to the angular momentum. m is the proton mass. When the nucleus is placed in an external magnetic field, Ho, the field exerts a torque upon the nucleus in— ducing precession of the Spin axis about Ho‘ The energy‘of a nucleus in the field is W = ’SI’iMHo' There are (21+l) quantized orientations and energies of the nucleus in the H2 " f" spin = - ' mks II I} NIP n»> lit 2 run an: 'h:” 35: : n i I! I5: Spin =- +- (b) Fig. l. Orientations (a) and energy levels (b) for 1=% Iota external magnetic field corrSSponding to the allowed values I, I-1, ..., 41 of the magnetic quantum number, M. For the proton (I=%), illustrated in Figure 1, there are two energy levels. The frequency separation of the energy levels (W = 1.%SI’$HO) is equal to the precession frequency and is directly pr0portional to the field strength Ho; hezni ht 2- SIF‘Ho but, = 5‘ 81 a. hence, hi 3: 1.1%}... H and, 21H) a: (do so, 0. == ”Ho (eq. 1) Transitions among the nuclei in the different energy levels are induced by a small oscillating field applied to the sample in a radiofrequency coil perpendicular to Ho. A block diagrnm of the apparatus is pictured in Figure 2. When the radiofrequency is equal to that of the nuclear preces- sion about H0,resonance absorption occurs. The effect is equivalent to the oscillating field tilting the nuclear mag- -net with reSpect to the external field, thus changing the nuclear orientation and energy. This induces an emf in the receiver coil through which the change taking place can be observed. The resonance frequency of a nucleus is not a constant, but varies according to the compound in which it is present. This is due to the fact that the field at the nucleus is not the same as the applied external field. The paired electrons of the molecule in which the given nucleus occurs give rise to a diamagnetism and alter the effective field at the nu- cleus. Thus, one must take into account the electronic Shielding when calculating the field at the nucleus. Hnucleus -.-. Happlied' O’Hnucleus q“ is the shielding constant characteristic of the surround- ings of the given nucleus. ' This shift in the resonance frequency is called the chemical Shift. It is now frequently given the symbol,‘ , Tube with Sample Super-stabilizer Super-stabilizer T and Slow Sweep w and Slow Sweep F Transmitter Coil Magnet a) Pole Receiver Magnet . Coil P°1e \A Ra o a Receive Rad o Transmitt r I 6 Indicat}r ,Fig. 2. Simplified apparatus for basic NMR.exPeriment (5) diff« ma . 'Y t Spin hers I ‘ l where 5 is defined by the equation 5 = Ha - Hp x 106 (eq. '4) Hr - Hc is the field at the nucleus being measured and HP is the field at the reference nucleus. The chemical shift for the reference compound is chosen as zero. Most shifts are mea- sured at a constant frequency. Each group of non-equivalent nuclei within a molecule usually absorbs at a different frequency. Therefore, the chemical shift can be used for structural determinations as well as determination of chemical shielding. The chemical shift is greater for those nuclei which are more shielded. In addition to the absorption peaks resulting from the different chemical shifts of non-equivalent nuclei, there may be a fine structure due to the magnetic coupling of the spin of a given nucleus with the nuclear spins of its neigh— bors through the bonding electrons. This is called the elec— tron—coupled Spin-Spin interaction. The magnitude of the coupling is governed by the bond type so that valid infer- ences can be drawn about the character from the observed splittings. The resonance peak of one set of nuclei is split up into a number of peaks by a non-equivalent group of nu— clei in the same molecule. The number of component peaks is .equal to 2MI+l, where MI equals the total epin of the non— equivalent nuclei in the other group. alectron-coupled spin— spin interactions do not give rise to observable effects for structurally equivalent nuclei. The splitting is usually greatest between directly bonded nuclei, but may be trans- mitted through several bonds. It is.independent of the mag- ‘netic field and temperature. ' The energy of the Spin—Spin interaction is preportional to the dot product of th two nuclei, E = Jijfi'fj (6; (eq. 5) where J13 is the constant of prOportionality in cycles per second and is called the Spin-spin coupling constant. It is simply a measure of the electron-coupled Spin-Spin inter- action. . ’ 'For a few simple, rigid,’symmetrical molecules and for those molecules for which the rapid internal rotation effect- ively increases the symmetry, J is effectively equal to the multiplet separation in cycles per second. For more complex cases where J and ‘are of the same order of magnitude, there .are a number of methods for the calculation of the coupling constant, all of which are fairly complex themselves. The method develOped by Anderson and McConnell (7) is the most adaptable. It is an adaptation of the procedure outlined by Van Vleck (8). They showed how to calculate the spin—spin coupling constants and the internal chemical shifts , from a knowledge of the intensities and separations of the multiplet peaks. HISTORICAL ll III HISTORIClL The only work reported with gases under a high pressure has been done by Gutowsky and McClure (9). They built a brass pressure probe to contain samples for the nuclear magnetic resonance spectrometer. The only simple hydrocar- bon whose spectrum was obtained is methane. However, in two articles conflicting values for the chemical shift of me- thane are given. The values given are .3a3 (10) and 9.4 (ll), relative to water in both cases. Gutowsky defines the chemi- cal shift as J 2 Hr 1; He X105 (eq. 4) I" so his values for the chemical shift must be multiplied by {-10) in order to conform with the definition given in equa- tion 2. The values given above conform to Equation 2. Meyer, Saika and Gutowsky (12) have measured the chemi— cal shift of a'large number of substituted hydrocarbons with reference to water. When they took the average value of these chemical shifts they obtained some values which are in good agreement with the values reported later in this work. These average values along with some selected values of the chemical shifts they measured are given in Table I. Ogg (13) obtained the nuclear magnetic resonance Spec— trum for both gaseous and liquid prOpane. Re did not measure the chemical shift. He Was interested in finding out if there was a proton absorption shift between the spectrum of 12 the liquid and gas. He found none, but this was probably due to poor resolution. The Spectra of both gaseous and liquid prOpane showed only one peak. The Spectrum of the gaseous sample showed a large noise-to-signal ratio indicating a low sample.concentration. TABLE I PROTON CHEMICXL SHIFTs* OF SUBSTITUTED HYDROOIRBONS _ggmpound or group JH H20 0 CHE-C +4.1 02H50H +4.0 CH340= +3.3 , CHB-CHO +3.4 CH3-C +3.4 CHEF 4 +2.} CH3Cl +2.3 CHBBr +2.1 CHBI +1.8 CH2: -O.5 CH +2.4 C6H5—C CH +2.5 C=CH—C -O.8 66HSCH=CH2 +0.5 wThe reference compound is water and I is defined by Equation 2. . 13 Meyer and Gutowsky (11) have measured the chemical shift of the halomethanes with reSpect to methane and their values are shown in Table II. TABLE II PROTON CHEMICAL SHIFTS OF THE HALOMETHXNES Compound ~ {4' cs4 O on}? +0.71 CH2F2 +0.88 CHE} +1.03 CHBCI +0.71 CH2012 +0.99 CHCl3 +1.18 CH3Br +0.73 CHQBr2 +1.01 CHBr3 +1.24 CHBI +0.76 CH212 ' +0.99 CHI3 (cgii'n) +0.65 4? is, defined by Equation 4. Shoolery (14) has measured the chemical shift of some substituted ethanes with respect to ethane. He derived an empirical formula for the correlation of proton magnetic resOnance chemical shifts with electronegativities of substi— tuents. This formula Electronegativity = 2.1 + 4.5 (5m: '- 5“, ) l4 (eq. 5) The proton chemical shifts of the substituted ethanes are given in Table III. TABLE III PROTON CHEMICXL SHIFTS 1ND ELECTRONEGATIVITIES OF SUBSTITUENTS IN SOME SUBSTITUTED ETHANES Compound {“3 - 5: Electronegativity ... Pauling g7 glectronggativity _ 02H6 0.00 0.00 2.10 2.1 CQHBSH 0.00 0.12 2.64 2.5 CQHSI 0.05 0.18 '2.68 2.5 cznssr 0.05 0.21 2.82 2.8 CZHSNH2 0.00 0.17 2.86 3.0 02H501 0.02 _O.22 3.00 3.0 C2H50H 0.01 0.25 3.18 3.5 '*J is defined by Equation 4. More recent Work along this line has been done by Dailey and Shoolery (15). They studied the chemical shifts in the NMR spectra of a number of methyl and ethyl deriva— tives in an effort to find out how the electronegativity of an atom changes when it forms a part of different substi- tuent groups. The value of (ACH3 - ACH2) for the various ethyl halides was plotted against the values of the electro- negativity of the corresponding halogen as given by Huggins (16). A straight line resulted which obeyed the following 15 equation: Electronegativity = 0.02315(ACH3-ACH2) + 1.71 (6:61. 6) The observed values of (A CH3-ACH2) along with this equation were used to derive the relative electronegativities of the substituent groups. The linear relationship found for the halogens was assumed to hold for all substituents. TABLE IV THE ELECTRONEGlTIVITIES OF HALOGENS IN THE ETHYL HALIDES J -:—— ' a Group A CH3 - A CHL Electronegativity ‘- -I 42 3 2.68 -Br 53 2.94 -Cl . 64 3.19 -F 96 3.93 gr Electronegativity was found by the use of Equation 6. Tomita (l7) and Thomas, et, a1. (18) have studied the NMR spectra of solid methane. Tomita studied the phase dia- gram of solid methane and Thomas, at, 31, studied the varia— tion of the spin-lattice relaxation time with temperature. l.v;.....v_.m_._._ ., - , _ EXPERIMENT :XL 17 IV EXPERIMENTAL The nuclear magnetic resonance spectra of the compounds reported here were all observed with a model V-4300—2 Varian NMR spectrometer at a constant frequency of 40 megacycles per second. a magnet with 12“ pole pieces was shimmed for high resolution work and a superhyperstabilizer used to im- prove the resolution. The chemical shifts of the gaseous compounds were all measured using hydrogen as the reference compound by simultaneous observation of the hydrogen peak and those of the compounds studied. The chemical shifts of the liquid samples were measured with respect to benzene by substituting a benzene sample for the one being studied while maintaining a constant field. The chemical shift of hydrogen with respect to benzene was found. This was done by measuring the separation of the of the hydrogen and benzene peaks from the CH3 group peak of prOpene since this peak was observed in the spectra of both gaseous and liquid prepene. All of the chemical shifts were then referred to benZene. The chemical shifts of the proton signal from the reference signals were measured by the side band technique of Arnold _and Packard (19) by use of a calibrated audio oscillator. The gaseous samples were contained in a specially made gas cell (Fig. 3) which held the gases up to a pressure of 800 lbs./in.2 without bursting. The pressures were measured With an Ascroft diaphragm pressure guage. The pressures of 4n ./ I W I l ' j n I t g i. I 3 l I I I l I l I I ' l I Constructed of I I 3 5"! Teflon and wound ——-e: 16 r— with 3 turns of I I #30 enameled : : c0pper wire.-'*"/‘q_ , I I I I l L---J __L 1" Ly i Jack h 1e‘ 3' 1‘ ° 3"“ 2: Fig. 3. Diagram of gas sample cell. 18 l9 l4? Hole for COpper lead- in tube. ‘ 1 . I; [ I I 1 1 I : 2 t I 1 I r I --J.--.L..:L.' 3‘ r E 2r l A.“ I I . L—:4 M‘ Brass stock F—lm‘q ‘——" Fig. 4 a. Male part of pressure cap for gas sample cell. 16 threads to the inch _ K e _ - 'g/ :I I i-...-----...--.§ __I 1 ZO'LI'Jb—lé‘ F————-l O:IED‘——W Aluminum'stock Fig. 4 b. Female part of pressure cap for gas sample cell. 20 sample and of the reference gas were approximately equal in each case neglecting deviations from the ideal gas law. In those cases where the spectra of the sample gases were ob- served without any reference gas present, the pressure of the gas in the cell was equal to its vapor pressure. All the gaseous compounds which have a sufficiently low vapor pressure were condensed, at liquid air tempera- tures, into a 5mm 0.D. glass tube which was then sealed. The spectra of these liquid samples were then observed. gurity of Materials’ All of the samples were purchased and with one excep- tion were not purified further. a sample of prOpylene was deoxygenated to determine if oxygen has any effect on the line width of the Spectra. The prOpylene was deoxygenated by repeated condensation at liquid air temperatures followed by warming in vacuo. The deoxygenated sample was then sealed off in a %" 0.D. medium-walled glass tube and its Spectrum was observed. The pressure of the gas was about ten atmos- pheres which is its vapor pressure; the pressure of the gas when it was contained in the Teflon gas cell was about the same. The size of the sample was about the same in both cases. The boiling points, vapor pressures and purities of the samples are given in Table V. Source of Materials The following compounds were purchased from the Mathe— son 00., East Rutherford, N. J.; methane, ethane, ethylene, TXBLE V BOILING POINTS, VIPOR PRESSURES AND PURITIES or SiMPLfiS USED Compound Boiling Vapor % Purity Point Pressure r ( cl__ {PSI at 20°C) __ Hydrogen -255 2000* 99.8 'Methane -161.4 1500'X 99 Ethane - 88 528 ‘ 95 Ethylene -lO3.9 1250* 99.5 icetylene - 88.5 633.5 99.5 CyclOprOpane - 32.9 75 99.5 PrOpane - 42.2 109 99 Allene - 32 100 -- Propylene - 47.7 139 99 PrOpene-Q-d - 47. 139 99 Eutene-l - 6.3 23 ‘ 99 Traps—2-butene + 1.0. 15.2 99.3 gig—2—butene + 3.6 12.6 99.7 P.,yg,.s 5 '_?._:, 7" *Tank pressure cyclOprOpane, pr0pane, propylene, propyne and l-butene. Xcetylene was purchased from the Columbia Organic Chem— ical Co., Inc., Columbia, S. C.. PrOpene-2-d was purchased from Merck and Co., Ltd., Montreal, Canada. gi§-2-butene and traps-2-butene were purchased from the Phillips Petroleum Co., Bartlesville, Okla. 22 Hydrogen was purchased from the Ohio Chemical and Sur- gical Equipment 00., Cleveland, Ohio. RESULT S _24 V RESULTS The nuclear magnetic resonance spectra of the gases studied are shown in Figures 5 to 11. The reference gas is hydrogen in each case. The NMR spectra of the liquids stud- ied are shown in Figures 12 to 19. The reference in each‘ case is benzene. The chemical shifts are indicated on each figure as well as in Tables VI and VII. The spectra of the mixtures of gaseous hydrocarbons with hydrogen show only two peaks. With the exception of prOpylene, one of the two peaks is due to hydrogen and the other is due to the sample. The propylene spectrum (Fig. 11a) shows the internal chemical shift. \ctually since there are ‘ three nonpequivalent proton groups within the prOpylene molecule, three peaks should appear in the spectrum. ilso the spectrum for prOpane (Fig. 10 a) should show two peaks since there are two non-equivalent proton groups within the jprOpane molecule. The spectra are much more complex for both «af these compounds in the liquid phase where not only the ithernal chemical shift appears, but also the fine structure due to spin-spin coupling. _ Much higher resolution is obtained in the spectra of 'the liquid phase samples than in those of the gas phase. This can also be seen from the line widths. The peaks are much broader in the spectrum of a substance for the gas phase than they are in the spectrum for'the liquid phase. It was thought tduat this line broadening might be due to the presence of a 25 paramagnetic gas such as oxygen in the sample. a sample of prOpylene was deoxygenated and it was found that this had no effect on the line width. It is possible that some of the ‘ broadening results from saturation of the nuclei since ra- ther high RF power must be used with the gaseous samples. Because the resonance peaks of hydrogen and ethylene were too close together to be resolved, methane was used as the reference gas. Two reference gases were used with pros pylene. When hydrogen was used, peak i of the propylene spectrum was strengthened (Fig.11 b); when acetylene was used, peak B was strengthened (Fig. 110). Since the separa- tion of the two reference gases is 137 cycles per second and the separation of the two peaks in the gaseous propylene spectrum is 133 cycles per second, it is obvious that the peaks of the reference gases coincided with the sample peaks and that the chemical shift of peak a with respect to hydro- gen is nearly zero. The electronegativity of the CH3 group in propane has been plotted in Figure 20 with the electronegativities of the halogen groups in the ethyl halides as given in Table IV. Each of the spectra, Fig. 5 - Fig. 19(c), represents ab- sorption of radiofrequency radiation (43.0 m3)versus mag- netic field strength. Only differences in the latter are used and these are expressed in cycles_per second making use of the relation 1’ (an!) :3 {if ' H , 184 cps I Base line Figure 5. Spectrum of H2 and CH4 gas mixture. 176 0P5 Base line J Figure 6. Spectrum of H2 and 02H6 ' gas mixture. 212 CPS H CH4 Base line Figure 7. Spectrum of the 02H4 and CH4 gas mixture. 27 137 Ops H H2 CQHE Basegline Figure 8. Spectrum of the H; and CgHg gas mixture. lg ' .v“ 11’ .‘-L 175 2”! cps i H ‘ Cyclo- Base line B -— ”:9- "-» “1“- Figure 9. Spectrum of the H2 , .‘vr— M’A and cyclOprOpane gas mixture. -«— “-u—c—r’. J 1+ 4.4 ~ ~ —,. 1 , {Base line - *m‘. wanna-s -- A —--. -—.9 w--- Figure 10 a. Spectrum of prOpane gas. *‘IIF- 2' new I I 5 I l 28 Base line 128 Cps ,——w—- I PrOpane ' Figure 10 b. Spectrum of H and prOpane gas m xture. V A B I " 1“: .Base line Figure 11 a. Spectrum of propylene gas. 130 Ops , . Base line "1 I” All”- Figure 11 b. Spectrum of H2 and propylene gas mixture. The H peak falls on peak g of prOpylene. 130 Ops e se Figure 11 c. Spectrum of acetylene and propylene gas mixture. 29 The acetylene peak falls on peak B of prOpylene. A B C DEFG Baserline Figure 12. Spectrum of liquid p ropa ne e J: 2.5 k ac line Figure 13. spectrum of liquid allene. 'R- 134 eps w—é QC: CH3 Figure 14 a. Spectrum of liquid prOpylene. 4.6 cps_ A I Base line A * B Figure 14 b. Spectrum of the chzcs— ’ group of prOpylene. 5.5 cps ——vI J. se line k Figure 14 c. Spectrum of the CH3 group of pr0pylene. H— 128 Ops ——II A B { Hg = *‘ -CHL Figure 15. Spectrum of liquid prOpene-2—d. (r- 5.25 n Base line Figure 16. Spectrum of liquid prOpyne. 31 32 it ch=cu- -CH2— mom.S Figure 17 a. Spectrum of liquid butane-1. 3 cps fir— A JV;\~ Figure 17b. Spectrum of H2C:CH- group of butene—l. se line -CH9- -CH3 Figure 17 c.Spectrum of CH2 and CH3 groups of butane-1. 33 ~re——150 cps —-| . j. -CH--v ~CH3 Figure 18 a. Spectrum of liquid trans-2-butene. A Base line A Figure 18 b. Spectrum of -CH; group of trans-2-butene. 7.5 chW—fi J . Figure 18 c. Spectrum of -CH3group of trans-2-butene. Base line v 153 Ops j .24: _ , =CH- -CH3 Figure 19 a. Spectrum of liquid cis—2—butene. 4 Ops a ‘__ Base line Figure 19 b. Spectrum of the =CH- group in cis-2—butene. ase line Figure 19 0. Spectrum of the CH3 group in cis-2-butene. TXBLE VI CHEMICXL SHIFTS OF GASEOUS HYDROCXRBONS 35 ~:1:: Chemical shift . Gas Separation from the benZene peak (cpslg. 1 Hydrogen 93 2.325 Methane 277 6.925 Ethane 269 5.725 Ethylene '65 1.625 Acetylene 230 5.750 Propane 221 5.525 CyclOprOpane 268 6.725 PrOpylene Peak a 93 2.325 Peak B 223 5.575 if :is defined by Equation 2. A!» II ,1‘ r.l ,V, l2..‘\ . . ek 1“: I12“ .‘ 1.93 ”1‘ 'i . Y.” Flaw. r n {I .I n TKBLE VII CHEMICAL SHIFTS OF LIQUID HYDROCARBONS 36 Liquid Separation from the {*5 A J benZenegpeak (cps) (cps) (cps) illene 100 2.500 Propane Peak 1 240 6.000 Peak B 245 6.125 14-58 2'75 (a) PrOpylene Peaks 1&B 92 2.325 Peak 0 226 5.650 134'0 5'5 (b) PrOpene—2-d Peak i 120 3.000 Peak B 248 6.200 130°C Propyne 210 5.250 Butene-l Peaks &&B 80 2.000 Peak C 200 5.000 7 5 ( ) Peak 0 245 6.125 - a is-2—butene Peak A 67 1.675 153 0 5.5 (b) Peak 3 220 5.500 ° 0.8 (c) Trans-2-butene Peak a 70 1.750 150 0 Peak B 220 ’ 5.500 ' (a) JCH2_CH3 (b) J H-CH bond) (attached to the same carbon atom of the double (c) JH-CH3 (attached to different carbon atoms of the * double bond) J'is defined by Equation 2. 4.._. Electro— negativity 31- L I P Fig. I 4 J 1 l l l IO 20 30 40 ACHB 20. Plot of electronegativity of substituent 50 6O —ACH2 giakni . I 70 80 90 100 57 versuSHACH3-60H2 for substituted ethanes. C lLCUL 8T IONS Vi - VI CiLCULiTIONS The method used in calculating the spin-spin coupling constant and internal chemical shift of pr0pane is taken from the work of inderson and McConnell (7). TiBLE VIII RELXTIVE INTENSITIES 1ND SEPXRXTIONS OF THE PROPANE PEXKS Peak Relative intensities Distance of peak (IQ from center line A 29 14.5 cps B 44 9.5 Ops C 140 2.5 cps D 98 2.5 cps E 24 5,5 cps F 14 7.5 Ops G 48 - 8.5 cps The Spin-Spin coupling constant (J19) can be calculated from the folowing equation; (J >2 = 3 “AW"? - W 12 21( I+1)(N1+N2T ((4407:) N1 2 N1+N°2) The internal chemical shift (A) can be calculated from the following equation; 2 (A) = NilN;1( N1+N2) 2 N1 is the total number of hydrogen nuclei in the CH3 ,la 40 groups_(six), N2 is the total number of hydrogen nuclei in the CH2 group‘(two), and I is the spin of the hydrogen nu- cleus (one-half). ‘ (“9"“) and (“a“)z) can be calculated from the fol- lowing relationship; m 4‘ «WV = (KW-wk» (we) is the observable mean resonance frequency and is equal to zero because the reference point was chosen half— way between the subgroups. Therefore, ‘1. <(Aw‘7m> z: ((0) > 99 +N+H+49 one + ’1’] + 2? + memo) + mflvm (17.5)2 (2 9) 9arzvrn+qsuvo 291129 * ((4 «~99 :- 4331’s; (lac-ml) =- nas (m)?- (4:09" : iflw‘” L‘ < ) {v ((4099 : c2.5)f‘(?8)+(5.5)"(29)+(25)“(H)+(8.5)"(qa) 781‘2‘! fI‘H‘mf-Ho +qq+zq - 12.52%») + (4.5mm m. 5) "( 2 9) 98 +21 # I? 1 78+190+97+ 2? 41 _§9‘6,97 ((409.9 :' 70148— <(Aw‘)" .2. 99oz.» Len)" The value of the spin-Spin coupling constant then is; (1:2)?- ; 3 ((4099 _ (~,’m§)((aw)1) 1 21(190)(N,+Na) (“U-)2.) M”; (MM/flj (J. )z 2 -..—-3 W 31903.50 ' - (animus; l I (h!) (6+2) 39.8 6 (I2) (8) (1..)2: Ar, (123.02 - 92. 87) I. The value of the internal chemical shift then is; (4)1 .~. “1’33” ((awv‘) (‘)2 1 +2,- (39.") (A)2 = 202.9; A 2 19.85" (CPS) 1+2 Calculation of the electronegativy of the CH3 group in propane by use of Equation 6. Electronegativity 0.02515(ACH3-40H2) + 1.71 0.02315(14.58) + 1.71 II N 0.34 + 1.71 2.05 ll DISCUSSION 44 VII DISCUSSION The nuclei which are more shielded are shifted further to the high field side of the Spectrum; iLgL, have larger positive values of the chemical shift (Eq. 2). The more pro- tons there are grouped around a carbon atom the better shielded these protons will be. On this basis then, methane should have the largest chemical shift of the compounds' which were studied. This is so. The chemical shift of me- thane with reSpect to benzene is 6.725. The magnetic field in all the spectra shown increases from left to right, so the chemical shift (defined by Eq. 2) also increases from left to right. Therefore, in the spectra shown, the peak on the extreme right is due to the group which is most shielded. If one went on this basis, one might expect the proton in the CH group to show the least chemical shift. However, this is not so. The chemical shift does not show a consist- ent decrease in going from ethane to ethylene to acetylene. The shifts for these compounds are 6.725, 1.625, and 5.75 reSpectively. The acetylene molecule which has two ECH groups has a larger chemical shift than the ethylene mole- cule which has two =CH2 groups. The reason for this effect is the multiplicity of the bonds attached to the carbon nu- cleus. The I! electrons in the triple bond rotate about the carbon-carbon axis under the influence of an external mag- netic field resulting in a Larmor precession which gives riSe to an added field at the nucleus (20). 45 The size of the molecule also affects the chemical shift. The larger the molecule the less the nuclei are shielded and the smaller is the chemical shift. This can be. seen in the three compounds methane, ethane and pr0pane. The chemical shift decreases from methane to propane. The reSpective values of the chemical shift are 6.925, 6.725 and 5.525. However, no quantitative predictions can be made of the value of the chemical shift from either the size or the electronic shielding of the molecule. Cy010pr0pane, because of its ring structure is in a class by itself. Its chemical shift is quite high (6.75), signifying that its protons are as well shielded as are the protons in ethane even though cyc10pr0pane consists of CH2 groups while ethane consists of CH3 groups. The C-C bonds in cyclOprOpane are shorter than a usual 1 bond indicating some double bond character. This shortening of the C-C bond in- creases the shielding of the protons and increases the chem- ical shift. Pr0pane and pr0py1ene were the only two gases capable of showing fine structure which were observed as gases. The vapor pressures of the other substances are too low to give a high enough concentration of nuclei in the gas phase. The other gases — allene, propene-2-d, pr0pyne, l-butene, gig-2- butene and tying—2—butene were condensed into a 5mm 0. D. glass sample tube and observed as liquids under a pressure equal to their vapor pressures. Propane and pr0pylene were 46 also observed as liquids and their liquid Spectra gave a much higher resolution than their gaseous spectra. The Spectrum of liquid propane (Fig. 12) consists of a single group of multiplets arising from both the CH3 and CH2 ‘protons. The fine structure shown is due to the spin-Spin coupling between these two groups of non-equivalent protons. The spectrum of liquid allene shows one peak only. The protons are all equivalent. The chemical shift is 2.5 which indicates that the allene protons are better shielded than the ethylene protons. The spectrum of liquid pr0pylene (Figs. 14 a to 14 c) shows three groups of multiplets. However, two of the groups of peaks, those due to the CH and CH2 groups, fall on one— another and are hard to differentiate. This group of peaks is characteristic of the vinyl group (CH2=CH-). The vinyl group is also present in l-butene and this same character- istic spectrum appears. The third group of peaks is due to the CH3 group. This group of peaks is a doublet indicating that there is no Spin-Spin coupling between the protons in the CH3 group and the protons in the CH2 group. The CH2 group also gives a doublet peak. Thus, both of these groups of protons interact only with the protons in the CH group. The only proton spin-Spin coupling effects transmitted a- cross the double bond are.thase of atoms attached to the carbon atoms of the double bond. The spectrum of liquid propene-2—d (Fig.15) also bears 47 out this premise. There are just two single peaks present. The peak on the left of the Spectrum is due to the CH2 group and the peak on the right of the Spectrum is due to the CH3 group. There is no Spin-Spin coupling between the protons of the two groups. The effects are not transmitted across the double bond. The Spectrum for liquid propyne (Fig.16) apparently shows only one peak. The peak due to theHICH group is ob- scured by the peak due to the CH3 group. The shielding power of the triple bond is evident here. The CH group has about the same chemical shift as the CH3 group. The effects of Spin-Spin coupling are not transmitted across the triple bond either so there is no fine structure. The Spectra of the butenes are quite complicated. There are several non-equivalent groups in the molecules and they interact to produce Spectra which are hard to interpret. The Spectrum of’lebutene (Figs. 17 a to 17 c) consists of four groups of peaks. Two of these peaks are the charac- teristic peaks of the vinyl group. These are on the left side (low field side) of the Spectrum. The CH3 group, which is the best shielded, is on the right or high field Side of the Spectrum. The remaining peak is due to the CH2 group ad— Jacent to the CH3 group. In Figure 17 b the Spectrum of the vinyl group is shown. It is a little different from the Spec— trum of the vinyl group shown in Figure 14 b. This is be- cause the CH group in l-butene has more groups to interact 48 with than does the CH group in pr0py1ene. The prominent doublet in Figure 17 b is due to the terminal CH2 group. The protons in the terminal CH2 group interact only with the proton in the CH group. The remainder of the Spectrum in Figure 17 b is due to the proton in the CH group. There is a large amount of fine structure which is partially obscured by the CH2 peak. Both the CH3 and CH2 peaks show some fine structure (Fig. 17 c). The Spectra of the 2—butenes are somewhat alike in that they both Show two groups of peaks about the same distance apart. The taggs compound (Figs. 18 a to 18 0) shows more fine structure in both groups of peaks than does the gig compound (Figs. 19 a to 19 c). For both compounds the group of peaks on the low field side of the Spectrum is due to the CH groups and the group of peaks on the high field side of the Spectrum is due to the CH3 groups. The peak for the =CH group of gig-2-butene (Fig. 19 b) is a simple quartet indicating that the protons on the ethylenic carbon atoms are only affected by the protons of the OH; group on the same side of the double bond. The spin- fl 0 '0 /q I" ‘ “ ”‘c ”’C\ /C: I ~\ / ‘1‘ I a H I fl 4! ”,C~u R. cis—2-butene trans-2-butene spin coupling effects are not transmitted across the double bond. However, the Spectrum of the =CH group of trans-2- butene (Fig 18 b) is composed of several peaks indicating 49 that the proton in each =CH group is affected by both CH3 groups. This seems to indicate tthat the Spin-Spin coupling effects can be transmitted across the double bond if the two interacting groups are gig to one another. For gig—E-butene the methyl peak is essentially a dou- blet indicating that the protons in each CH3 group interact with the proton in only one =CH group. In Figure 18 a the peak for the CH3 groups seems to be a quadruplet indicating interactions with both =CH groups. However, Figure 18 c, which shows the CH3 peaks of tggns-2-butene under a higher resolving power, shows a larger number of peaks. These peaks are probably caused by interactions between the CH:5 groups. The CH groups when interacting with the CH3 groups cause the CH3 groups to interact with each other. These interactions are weaker than the primary interactions. However, they serve to complicate the Spectrum. The propane molecule has been treated as if it were a substituted ethane. The electronegativity of the CH3 group has been calculated by the method of Dailey and Shoolery (15). The electronegativities of some substituent groups of ethane have been plotted against the values of (ACH3 -ACH2). This plot is shown in Figure 20. The electronegativity of-the CH3 group has been determined to be 2.05 (see calculations, page' 42) using the eXperimental value cf (OCH3 -ACH2) for the CH3 group found from the prOpane Spectrum. SUMM XRY 51 SUMMXRY The nuclear magnetic resonance Spectra of simple gas— eous and liquid hydrocarbons were observed under high resolu- ‘tion by use of the model V-4300-2 Varian NMR Spectrometer system. A Special container was constucted of Teflon to con- tain the gaseous samples under pressures up to 40 atmOSpheres. Liquid samples were contained in sealed pyrex tubes. The chemical shifts (K) of the peaks in the Spectra of liquid propane, pr0pylene, propene-2—d, propyne, allene, and the isomeric butenes were measured with reSpect to benzene and those of gaseous ethane, ethylene, acetylene, propane, pr0pylene, cyclOprOpane, and methane with reSpect to hydro- gen gas mixed with the gas being studied. Where possible - nuclear Spin-Spin coupling constants (J) for the electron- coupled Spin-Spin interactions between protons in neighbor- ing groups have been obtained also. The Spectra have been interpreted in terms of the chem- ical shifts and Spin-Spin coupling constants. In the case of prOpane the chemical shift and Spin-Spin coupling constant were found by the method of Anderson and McConnell (7); from the former a value of the electronegativity of the CH5 group has been derived. REFERENCES 53 REFERENCES . Hertz, Chem. Rev. 55. 829 (1955). W (1) J. (2) H. . Gutowsky, Ann. Rev. Phys. Chem. 5, 333 (1954). J . S. Smith,'Quart. Rev. (London) 1, 279 (1953). E PO) (3) (4) .,R. Andrew, "Nuclear Magnetic Resonance," Cambridge university PreSs, Cambridge (1955). (5) “Radio-Frequency Spectrosc0py,' Varian Associates Tech- nioal Information Bulletin 1, No. 1 (1953). (6) H. M. McConnell, A. D. McLean and C. 1. Reilly, J. Chem. Phys. 23, 1152 (1955). (7) W. Anderson and H. M. McConnell, J. Chem. Phys. gé, 1499 (1957). (8) J. H. Van Vleck, Phys. Rev. Zfl, 1168 (1948). (9) H. 8. Gutowsky and R. E. McClure, Phys. Rev. 8;, 277 (1951). (10) H. S. Gutowsky and C. J. Hoffman, J. Chem. Phys..12, 1259 (1951). (11) L. H. Meyer and H. S. Gutowsky, J. Phys. Chem. 51, 481 (1953). ' (12) L. H. Meyer, 1. Saika and H. S. Gutowsky, J. Am. Chem. Soc. 15, 4567 (1953). (13 R. 1. 033, J. Chem. Phys.ng., 560 (1954). (14) J. N. Shoolery, J. Chem. Phys. 21, 1899 (1953). (15) B. P. Dailey and J. N. Shoolery, J. Am. Chem. Soc. 11, 3977 (1955). (16) M. L. Huggins, J. Am. Chem. Soc. 15, 4123 (1953). 54 (17) K. Tomita, Phys, Rev. eg, 429 (1953). (18) J. T. Thomas, N. L. Alpert and H. C. Torrey, J. Chem. Phys. 18, 1511 (1950). (19) J. T. Arnold and M. G. Packard, J. Chem. Phys. 12, 1608 (1951). ‘ (20) J. 1. Peple, J. Chem. Phys. 23, 1111 (1956). w. "iv. .tsTFY LIBR’Q'RY 7 J T116818 GEMISTRY LIBRARY c.2 Laine. Norman R. 31.8. 1958 AN STATE U 31293 IVERS TY Ll I BR 085 5476 ARIES MICHIG | Ill)” 03