ISTANBUL TECHNICAL UNIVERSITY GRADUATE SCHOOL OF SCIENCE ENGINEERING AND TECHNOLOGY
M.Sc. THESIS
JUNE 2013
SYNTHESIS AND CHARACTERIZATION OF POLYACRYLONITRILE-BASED CARBON NANOFIBERS
Ezgi İŞMAR
Nano Science and Nano Engineering
Nano Science and Nano Engineering Program
Anabilim Dalı : Herhangi Mühendislik, Bilim Programı : Herhangi Program
JUNE 2013
ISTANBUL TECHNICAL UNIVERSITY GRADUATE SCHOOL OF SCIENCE ENGINEERING AND TECHNOLOGY
SYNTHESIS AND CHARACTERIZATION OF POLYACRYLONITRILE -BASED CARBON NANOFIBERS
M.Sc. THESIS Ezgi İŞMAR (513111006)
Department of Nano Science and Nano Engineering
Nano Science and Nano Engineering Program
Anabilim Dalı : Herhangi Mühendislik, Bilim Programı : Herhangi Program
HAZİRAN 2013
İSTANBUL TEKNİK ÜNİVERSİTESİ FEN BİLİMLERİ ENSTİTÜSÜ
PAN TABANLI KARBON NANOELYAF SENTEZİ VE KARAKTERİZASYONU
YÜKSEK LİSANS TEZİ Ezgi İŞMAR
(513111006)
Nano Bilim ve Nano Mühendislik Anabilim Dalı Nano Bilim ve Nano Mühendislik Programı
Anabilim Dalı : Herhangi Mühendislik, Bilim Programı : Herhangi Program
v
Thesis Advisor : Prof. Dr. A.Sezai SARAÇ ... Ġstanbul Technical University
Jury Members : Prof. Dr. Nilgün KIZILCAN ... Istanbul Technical University
Prof.Dr. Nilgün Kızılcan Prof.Dr. Yücel Şahin
Prof. Dr. Yücel ŞAHİN ... Anadolu University
Ezgi İşmar, a M.Sc. student of ITU Graduate School of Science Engineering and Technology student ID 513111006 successfully defended the thesis entitled ―SYNTHESIS AND CHARACTERIZATION OF POLYACRYLONITRILE -BASED CARBON NANOFIBERS‖ which she prepared after fulfilling the requirements specified in the associated legislations, before the jury whose signatures are below.
Date of Submission : 03 May 2013 Date of Defense : 03 June 2013
vii
ix FOREWORD
First of all, I would like to express my gratitude to my thesis advisor, Prof. Dr. A. Sezai SARAÇ for his continuous patience, guidance and helpful critics in my studies. Special thanks to Prof. Dr. Karen de Clerck and Sam van der Heijden for their experiences shared with me.
In addition, I am thankful to all my colleagues and friends in this research ; Onur AYAZ, Burcu ARMAN, Dilek SUADĠYE, Selda ġEN, Serhan AYA, Leyla YAĞMUR, , M.Tolga SATICI, Özgür CEYLAN, BaĢak DEMĠRCĠOĞLU, Keziban HÜNER, Timuçin BALKAN, Diğdem GĠRAY, Mustafa Edhem KAHRAMAN, Nazif Uğur KAYA, Aslı GENÇTÜRK, Giray ERSÖZOĞLU, Ömer Faruk VURUR, Hacer DOLAġ and Ġlknur GERGĠN for their assistance, encouragement and friendship.
My last gratitude goes to my parents for being with me and supporting me at every moment of my life.
June 2013 Ezgi ĠġMAR
xi TABLE OF CONTENTS Page FOREWORD... ix ABBREVIATIONS ... xiii LIST OF TABLES ...xv
LIST OF FIGURES ... xvii
SUMMARY ... xxi ÖZET ... xxiii 1. INTRODUCTION ... 1 1.1 Purpose of Thesis ...1 1.2 Literature Review ...2 1.2.1 Carbon fibers ... 2
1.2.2 PAN-based carbon fibers ... 3
1.2.3 Electrospinning of nanofiber ... 5
1.2.4 Carbon nanofibers... 6
2. MATERIALS AND METHODS... 7
2.1 Materials ...7
2.2 Equipments and Analysis ...7
3. EXPERIMENTAL STUDIES ...11 3.1 Synthesis of Copolymers ... 11 3.2 Electrospinning Process ... 12 3.2.1 Solvent preparation ...12 3.2.2 Sample preparation ...14 3.3 Oxidation ... 14 3.4 Carbonization ... 14
4. RESULTS AND DISCUSSIONS ...15
4.1. Characterization of Polymers and Nanofibers ... 15
4.1.1. FTIR-ATR spectroscopy ...15
4.1.2 Morphology of electrospun nanofibers ...19
4.1.3 Differential scanning calorimetry (DSC) ...23
4.1.4 Thermal gravimetric analysis (TGA)...25
4.1.5 Dynamic mechanical analysis (DMA) ...25
4.2 Characterization of Oxidized and Carbonized Nanofibers .. Hata! Yer işareti tanımlanmamış. 4.2.1 FTIR-ATR spectroscopy ...30
4.2.2 Morphology of nanofibers ...36
4.2.3 Differential scanning calorimetry (DSC) ...39
4.2.4 Dynamic mechanical analysis (DMA) ...40
4.2.5 Thermal gravimetric analysis (TGA)...46
5. CONCLUSIONS AND RECOMMENDATIONS ...49
xii
APPENDICES ... 55 CURRICULUM VITAE ...Hata! Yer iĢareti tanımlanmamıĢ.
xiii ABBREVIATIONS
AA : Acrylic Acid ANF : Aligned NanoFiber CNF : Carbon Nanofiber
DMA : Dynamic Mechanical Analysis DMF : Dimethylformamide
DSC : Differential Scanning Calorimeter FNF : Flat NanoFiber
FTIR-ATR : Fourier Transform Infrared Spectroscopy-Attenuated Total Reflection
IA : Itaconik Acid PAN : Polyacrylonitrile Rpm :Revolution Per Minute
SEM : Scanning Electron Microscope TGA : Thermo Gravimetric Analyze
xv LIST OF TABLES
Page Table 3.1: Polymerization parameters...11 Table 3.2: Electrospinning parameters. ...13 Table 4.1: DSC-enthalpy values for different oxidation duration and temperature. .40 Table 4.2: Average Young’s Modulus values after oxidation. ...41 Table 4.3: % of weight loss after waiting 300 min. at different temperatures. ...47
xvii LIST OF FIGURES
Page Figure 1.1: Flow chart of production of carbon fibers from PAN precursor. ... 3 Figure 1.2: Schematic view of oxidation. ... 4 Figure 1.3: Schematic representation of nanofiber production with stable and
rotating collector. ... 6 Figure 4.1: FTIR-ATR spectra of PAN powder, film and nanofiber. ...16 Figure 4.2: FTIR-ATR spectra of P(AN-AA) co-polymer for different feeding of
wt% of AA. ...16 Figure 4.3: Correlation between FTIR-ATR absorbance ratio( CN/C=O peaks) with
feeding wt% of AA. ...17 Figure 4.4: FTIR-ATR spectra of P(AN-IA), feeding weight ratio of AN/IA, 93/7. 17 Figure 4.5: FTIR-ATR spectra of P(AN-AA), feeding weight ratio of AN/AA, 93/7.
...18 Figure 4.6: Weight ratio of PAN polymer inside the solvent vs. viscosity. ...18 Figure 4.7: SEM images of electrospun flat nanofiber webs; (a) 6wt% PAN, (b)
7wt% PAN, (c) 8wt% PAN - scale is 2 micrometer and (d) 9wt% PAN, (e) 10wt% PAN, (f) 11wt% PAN and scale is 5 micrometer. ...20 Figure 4.8: Average diameter distribution vs. wt% of PAN. ...21 Figure 4.9: Diameter histogram for 6 wt% PAN flat nanofiber web. ...21 Figure 4.10: SEM images of electrospun aligned nanofiber webs; (a) 6wt% PAN ,
(b) 7wt% PAN, (c) 8wt% PAN d (d) 9wt% PAN, (e) 10wt% PAN, (f) 11wt% PAN (scale is 2,1,1,2,2,5 micrometer, respectively). ...22 Figure 4.11: DSC curves for P(AN-IA), P(AN-AA) co-polymers and homo polymer PAN ...23 Figure 4.12: DSC compration between powder formation and nanofiber formation of
PAN. ...24 Figure 4.13: DSC curve for 8%wt of P(AN-AA) co-polymer, aligned nanofiber
formation. ...24 Figure 4.14: TGA comparison of PAN nanofiber web, film and powder. ...25 Figure 4.15: Stress-strain curve for aligned nanofiber web with average value and
multiple form. ...26 Figure 4.16: Stress-strain curve for flat nanofiber web with average value and
multiple form. ... 266 Figure 4.17: Tan delta curve from DMA measurement for investigating Tg value for P(AN-AA), 93/7 co-polymer, aligned nanofiber web with error bars. 277 Figure 4.18: Tan delta curve from DMA measurement for investigating Tg value for PAN homo polymer, aligned nanofiber web with error bars. ... 277 Figure 4.19: Comparison of PAN and P(AN-AA) aligned nanofiber tandelta curves
for Tg values... 368 Figure 4.20: Correlation between Young’s Modulus and wt% of PAN both for
xviii
Figure 4.21: Relationship between average fiber diameter and viscosity both for aligned and flat electrospun fibers. ... 299 Figure 4.22: Combination of Young’s Modulus and average fiber diameters. ... 29 Figure 4.23: FTIR-ATR spectra of oxidized PAN aligned nanofiber webs at 235 oC
for different durations. ... 30 Figure 4.24: FTIR-ATR spectra of oxidized PAN aligned nanofiber webs at 270 oC
for different durations. ... 31 Figure 4.25: Correlation between FTIR-ATR absorbance ratio of CN/C=O with
temperature. ... 31 Figure 4.26: FTIR-ATR spectra comparison between oxidation at 235 and 270 oC
during 30 min. ... 322 Figure 4.27: Correlation between FTIR-ATR absorbance ratio for different ... 332 Figure 4.28: Correlation between FTIR-ATR absorbance ratio for different
oxidation durations, oxidation temperature 270 oC. ... 333 Figure 4.29: Correlation between FTIR-ATR absorbance ratio for different
oxidation durations, oxidation temperatures 235 and 270 oC. ... 343 Figure 4.30: FTIR-ATR spectra of P (AN-AA), oxidation @ 270 oC for different
time periods... 344 Figure 4.31: Relationship between duration of oxidation and FTIR-ATR absorbance ratio for P(AN-AA) nanofiber web. ... 344 Figure 4.32: FTIR-ATR spectra of P(AN-AA) without oxidation and oxidation at
270 oC during 3h. ... 35 Figure 4.33: Color change after oxidation and carbonization from left to right;
oxidation at 235 °C-30 min.,oxidation 235 °C-2h, oxidation 300 °C-1h, Carbonization at 950 °C-1h. ... 366 Figure 4.34: 10% PAN ANF as precursor, histogram of fiber diameter and SEM
image. ... 36 Figure 4.35: 10% PAN ANF was used as precursor and (a) after oxidation at 270
C-1h, (b) after carbonization of (b), (c) after oxidation at 270 C-3h, (d) after carbonization of (c). ... 377 Figure 4.36: Fiber diameter change on sample 10% PAN ANF, after oxidation at
270 C- 3h and carbonization at 950 C- 1h. ... 388 Figure 4.37: Fiber diameter change on sample 10% PAN ANF, after oxidation at
270 C- 1h and carbonization at 950 C- 1h. ... 388 Figure 4.38: (a) 8% P(AN-AA) ANF, (b) after oxidation at 270 C-3h, (c) after
carbonization at 950 C-1h. ... 39 Figure 4.39: Fiber diameter change on sample 8% P(AN-AA) ANF, after oxidation
at 270 C- 3h and carbonization at 950 C- 1h. ... 39 Figure 4.40: Conversion-time plots at different temperatures. ... 40 Figure 4.41: DMA stress-strain curve for 8% PAN ANF, oxidation at 235 oC-2h. 411 Figure 4.42: DMA stress-strain curve for 9% PAN ANF, oxidation at 250 oC-3h. 422 Figure 4.43: DMA stress-strain curve for 8% PAN ANF, oxidation at 235 oC-30min. ... 422 Figure 4.44: DMA stress-strain curve for 8% P(AN-AA) ANF, oxidation at 270 o
C-2h. ... 433 Figure 4.45: DMA stress-strain curve for 8% PAN ANF, oxidation at 235 oC3h. . 433 Figure 4.46: DMA stress-strain curve for 8% PAN ANF, oxidation at 270 oC-3h. 444 Figure 4.47: DMA stress-strain curve for 11% PAN ANF, oxidation at 270 oC-2h
xix
Figure 4.48: DMA stress-strain curve for 10% PAN ANF, oxidation at 270 oC-3h and carbonization at 950 oC-1h. ... 455 Figure 4.49: DMA stress-strain curve for 10% PAN ANF, oxidation at 270 oC-1h
and carbonization at 950 oC-1h. ... 455 Figure 4.50: TGA curve for same isothermal time (300min) for different
temperatues. ... 446 Figure 4.51: Relationship between weight loss and temperature. ... 466
xxi
SYNTHESIS AND CHARACTERIZATION OF PAN-BASED CARBON NANOFIBERS
SUMMARY
Nowadays nano materials have a great interest due to their unique properties compared to the bulk material. Also carbon nano materials such as carbon nano wires, carbon nanotubes (CNTs), fullerene, carbon nano fibers (CNFs) have a huge interest. Nanofibers have a huge aspect ratio also; nanofibers have different several application areas in engineering field. Nanofiber production is not a new technology but, with development in characterization tools now it is possible to better characterize nano-sized fibres compared to 100 years ago.
This study has two main goals, first one is synthesis a co-polymer of polyacrylonitrile (PAN) and second aims is producing a carbon nanofibers out of homo-polymer PAN and co-polymer of PAN and compression between these two products. This study consists of two main parts. First part of it to synthesis a co-polymer of PAN, in this perspective, acrylic acid (AA), itaconic acid (IA) and acrylonitrile (AN) were used as a raw material for polymerization reactions and copolymers were synthesized by using ammonium persulfate (APS) as an oxidant. Polymerization medium was water and dimethylformamide (DMF) mixture. According to monomer weight, feeding ratio different co-polymerization reactions was carried under same co-polymerization conditions. Synthesized co-polymers were investigated by Fourier Transform Infrared Spectroscopy-Attenuated Total Reflection (FTIR-ATR) spectroscopy and their thermal behavior was examined by using Differential Scanning Calorimeter (DSC) and Thermal Gravimetric Analyzer (TGA).
Results show that, addition of monomers to get co-polymer was decreased the Tg value of the co-polymer of PAN compared to homo polymer. Oxidation and carbonization steps are energy and time consuming process and for decreasing the initiation temperature of oxidation reaction provides better control on the process and also helps to consume less energy.
On the other hand, commercially available PAN polymer solved in DMF and electrospinning solvents were prepared. After optimum parameters have been set, nanofiber webs have been investigated by using Dynamic Mechanical Analysis (DMA) for investigating mechanical properties of web also Tg values was determined by using DMA. In addition, homo polymer PAN and co-polymers of PAN polymers were electrospun. After electrospinning step, thermal oxidation and carbonization steps were applied to nanofiber webs to obtain carbon nanofiber web. Oxidation procedure was studied between temperature range 200-300 oC and for carbonization step; low temperature carbonization was applied under inert atmosphere at 950 oC for all samples.
xxii
Oxidized and carbonized samples were characterized by using DMA for their mechanical properties and for thermal properties DSC and TGA were used and for structural analysis, FTIR-ATR spectroscopy was used. For all steps, surface morphology of the fibers have been observed with Scanning Electron Microscope (SEM).
Consequently, PAN-based carbon nanofiber webs have been produced successfully out of homo polymer and co-polymer of PAN. Handling problem is the major problem while carrying and measuring the sample properties for carbonized and oxidized samples. Further studies should be focus to improve handling of the sample. For better understanding of the oxidation step, experimental studies can be reproduced for different oxidation temperatures.
xxiii
PAN-TABANLI KARBON NANOELYAF SENTEZİ VE KARAKTERİZASYONU
ÖZET
Nano malzemeler her geçen gün daha fazla ilgi görmektedir. Üstün özelliklerinden dolayı karbon nano malzemeler de yoğun ilgi görmektedir. Nanoelyaflar da bu alanın baĢında gelmektedir. Birçok uygulama alanları mevcut olan nanoelyaflar yüksek boy-en oranları sayesinde çeĢitli mühendislik uygulama alanlarında kullanıma uygundur. Karbon elyaflar polyacrylonitrile (PAN) tabanlı baĢta olmak üzere çeĢitli kaynaklardan katran ve selüloz gibi elde edilebilmektedir. PAN’ın kimyasal yapısında ki yüksek oranda ki karbon miktarı PAN tabanlı karbon elyaflarının kullanımının artmasına neden olmuĢtur. Akrilik elyaflarının içeriğinde %85 den çok PAN bulunmaktadır, akrilik elyafları karbon elyaf prekursörı olarak kullanmaktadır. Akrilik elyafları belirli miktarda polyacrylonitrile ko-polimerleri içermektedir. En yaygın kullanılan monomerlerin baĢında itakonik asit, akrilik asit, metil metakrilat ve vinyl asetat gelmektedir. Ko-polimer kullanımı iĢlenebilirliği az olan homopolimer polyacrylonitrile’ın iĢlenmesini kolaylaĢtırmasının yanı sıra karbonizasyona giden adımlar da enerji tasarrufu sağlamaktadır. Ko-polimer kullanılan durumlarda daha düĢük sıcaklıklarda çalıĢmak mümkündür ayrıca son ürünün mekanik özellikleri üzerinde olumlu etkisi de gözlenmektedir.
Bu çalıĢma kapsamında karbon nanoelyaf üretimi PAN tabanlı malzemelerden üretilmiĢtir ve süreç optimizasyonu yapabilmek adına önce homopolimer ile çalıĢıp daha sonra polyakrilonitril-ko-itakonik asit ve polyakrilonitril-ko-akrilik asit polimerleri çeĢitli besleme oranlarında sentezlenmiĢtir. Bütün polimelerin dimetilformamid (DMF) ile çözeltileri hazırlanmıĢtır ve bu çözeltiler oda sıcaklığında elektro çekim ünitesine beslenmiĢtir.
Elektro eğirme yöntemi ile mikro ve nano ölçekte nanoelyaf elde etmek mümkündür. Elektrik alan arasındaki voltaj farkına dayanan çalıĢma prensibi vardır, toplayıcı plaka ve besleme yapılan iğnenin uç kısmına yüksek voltajda gerilim uygulanır ve bu gerilimin etkisiyle polimer çözeltisi toplayıcı plakaya doğru yol alır ve çözücü malzeme büyük ölçü de buharlaĢır. Toplayıcı plaka üzerinde dokusuz kumaĢ yapısında elyaf tabakası meydana gelir.
Bu çalıĢma kapsamında PAN’ın kütlece farklı oranlarda DMF ile çözeltileri hazırlanmıĢ ve bunların elektro eğrilebilirlikleri kontrol edilmiĢtir. Elektro eğirme sonucunda elde edilen liflerin çapları morfolojik özellikleri ve mekanik özellikleri incelenmiĢtir.
Karbon elyaf üretim süreci son derece enerji yoğun bir süreç olup uzun soluklu bir iĢlemdir. Karbon nanoelyaf üretimi için de bu durum geçerlidir. PAN tabanlı nanoelyaf matları bu çalıĢma da karbon nanoelyaf prekursörü olarak kullanılmıĢtır.
xxiv
Homopolimer ve ko-polimerden elde edilen PAN tabanlı nanoelyaf matları önce oksidasyon sonra düĢük sıcaklık karbonizasyon iĢlemlerine tabi tutulmuĢtur.
Oksidasyon iĢlemi; 200 ile 300 °C arasında 30 dakika ile 3 saat arasında değiĢen oksidasyon süreleri kullanılarak, atmosfer ortamında yapılmıĢtır. Oksidasyon adımında PAN ın yapısında ki C≡N bağları açılarak C=N e dönüĢmekte ve C=O bağ sayısı artmaktadır. HalkalaĢma reaksiyonları ve kısmen dehidrojenasyon genaration reaksiyonları bu adımda gerçekleĢmektedir. Artan oksidasyon sıcaklığı ve süresi ile numune rengi beyazdan sarı, gün batımı rengi, açık kahverengi ve koyu kahverengi tonlarında değiĢmektedir. Oksidasyon iĢlemi sırasında numunede relaksiyasyon meydana gelir ve bunun sonucu olarak büzülme olur. Bu yüzden bu adımda numuneye çekme kuvveti uygulamak büyük önem taĢımaktadır.
Oksidasyon iĢlemini düĢük veya yüksek sıcaklık karbonizasyon adımı takip etmektedir. Yüksek sıcaklık karbonizasyon iĢlemi sonucunda yüksek young’s moduluslü karbon elyaf elde edilmektedir. Bu çalıĢma da düĢük sıcaklık karbonizasyon iĢlemi uygulanmıĢtır ve 950 °C sıcaklıkta 1 saat boyunca karbonizasyon iĢlemi gerçekleĢtirilmiĢtir. Son ürün olan karbon nanoelyaflar çok kırılgan bir yapıya sahiptir ve bu da karakterizasyon çalıĢmalarının zorlaĢmasına sebep olmuĢtur.
Üretilen nanoelyaf matlarının morfolojik özellikleri taramalı elektron mikroskobu ile incelenmiĢtir ve çaplar en az 50 adet elyaf ölçülerek hesaplanmıĢtır. Kütlece düĢük PAN konsantrasyonlarında kütlece %6 ve %7 değerleri için düzgün bir yapı gözlenmemiĢtir ayrıca kütlece %12’lik konsantrasyon da elektroeğirme yüksek viskozite değerinden ötürü greçekleĢtirilememiĢtir. Çap ile viskozite arasında artan bir trend gözlenmiĢtir.
Termal özellikler, diferansiyel taramalı kalorimetre ve termal gravimetrik analiz cihazı ile saptanmıĢtır. Oksidasyon süresi ve sıcaklığı arttıkça numunede ki kütle kaybı miktarı da artmaktadır.
Fourier dönüĢümlü kızılötesi spektroskopi- attenuaded total reflaectance (FTIR-ATR) cihazı sayesinde polimerlerin ve nanoelyafların kimyasal yapısı incelenmiĢtir. Oksidasyon iĢleminin zaman ve sıcaklık olarak nanoelyaf üzerindeki etkileri artan ve azalan piklerin oranlarından sıcaklık, süre ve oksidasyon iliĢkisini yansıtan grafiklere dönüĢtürülmüĢtür.
Karbon elyaf ve nano elyaf üretimi sürecinde çok fazla enerji harcanmaktadır. Ko-polimer ilavesi ile akrilonitrilin hem iĢlenebilirliğini kolaylaĢtırmak hem de camsı geçiĢ sıcaklığı (Tg) gibi malzemeye ait değerleri düĢürerek gerçekleĢecek reaksiyon sıcaklıklarını da azaltmak hedeflenmektedir.
Üretilen karbon nano elyaflar çok kırılgan malzemeler olup kullanım zorluğu bulunmaktadır. Karbonizasyon iĢleminden sonra malzemenin fırından çıkartılıp taĢınması ve ölçüm cihazlarına yerleĢtirilmesi kırılgan yapısından ötürü çok zor bir Ģekilde gerçekleĢtirilmiĢtir. SEM fotoğraflarından da görüleceği gibi elyafların bir kısmında kopmalar görülmektedir. Bu yüzden gerilme-germe Ģekil değiĢtirme eğrilerinden elde edilen değerler gerçek değerlerinin altındaki sonuçları yansıtmaktadır.
Karbonizasyon iĢlemi boyunca malzeme de %60’lara varan kütle kaybı yaĢanmaktadır. Oksidasyon iĢleminden sonra kütle kaybında önemli ölçüde bir değiĢiklik olmamakla beraber karbonizasyon iĢlemi ile kütle kaybı artıĢ
xxv
göstermektedir. Bu çalıĢma da TGA ve SEM ölçümlerinden elde edilen nanoelyaf çap değerleri kullanılarak süreç boyunca gerçekleĢen kütle kaybı hesaplanmıĢtır. Ġleriki çalıĢmalar ile numune taĢıma sorunu ortadan kaldırılabilir ise üretilen karbon nanoelyaf malzemeler kompozit yapıların içinde, çeĢitli mühendislik uygulamalarında kullanıma uygun bir hal alacağı düĢünülmektedir.
1 1. INTRODUCTION
1.1 Purpose of Thesis
In this study, Poly (acrylonitrile-co-acrylic acid) and Poly (acrylonitrile-co-itaconic acid) co-polymers have been synthesized in different weight ratios. After that, synthesized polyacrylonitrile-co-polymers and commercial homo-polyacrylonitrile analyzed by using Fourier transform infrared-Attenuated Total Reflectance spectroscopy (FTIR-ATR) and thermal behaviors analyzed using Thermo gravimetric analysis (TGA) and Differential scanning calorimetry (DSC). Then, electrospinnability of polymers checked for different parameters and dimethylformamide used as an electrospinning solvent. Oxidation and low temperature carbonization steps followed after producing the nanofiber webs and all mechanical properties of nanofibers examined by Dynamic mechanical analysis (DMA) and for all the steps surface morphology of the fibers observed with Scanning electron microscope (SEM).
This thesis has two goals, first one is synthesis to co-polymers of acrylonitrile and second goal is produce carbon nanofibers out of these polymers. Homo-polyacrylonitrile and co-polymers of acrylonitrile have been used for precursor for nanofiber webs then PAN nanofiber webs have been used for producing the PAN-based carbon nanofibers. In this perspective, FTIR-ATR spectroscopy, DSC and TGA results have investigated to compare, commercial polyacrylontirile and synthesized co-polymer behaviors. Effects of different PAN-based polymers on the nanofiber web have been investigated in terms of strength and fiber morphology by Dynamic mechanic analysis and Scanning electron microscope. Producing the neither convectional carbon fibers nor carbon nanofibers out of PAN source, stabilization/oxidation steps should be followed. Thus, in this study after producing the PAN nanofiber webs, oxidation step has been investigated for different oxidation parameters. For oxidation process, temperature has been changed between 200-300 °C and duration of oxidation time changed 30 min. to 3h. Results have been
2
investigated by FTIR-ATR spectroscopy and DMA. Oxidation step has been followed by low temperature carbonization process. Result of this study pointed that, produced carbon nanofiber webs can be used in engineering fields for instance, reinforcement material for different polymeric materials as a composite material to improve their mechanical properties.
1.2 Literature Review
In this chapter, detailed literature review presented which is related to the experimental studies within the thesis study.
1.2.1 Carbon fibers
Usage of carbon fiber has an old past and, was first patented as the filament in the incandescent electric lamp in 1880 (Morgan, 2005, pp. 65–68).
Carbon fibers have an excellent mechanical properties such as stiffness and high strength besides their light weight properties (Edie, 1998).
First PAN-based carbon fiber was introduced by Shindo of Osaka Industrial Institute of Japan in 1959 and detailed reported was presented in 1961 (Morita et al., 1986). It is possible to get carbon fibers out of PAN, pitch and cellulose sources (Edie, 1998; Morgan, 2005). Polyacrylonitrile and co-polymers of polyacrylonitrile are suitable precursors for carbon fibers (Bashir, 1991). PAN includes 67.9% of carbon content in its structure so it is not surprising to use polyacrylonitrile as a carbon fiber precursor (Morgan, 2005, p. 121)
The PAN precursor is a form of acrylic fiber and 90% of commercial carbon fibers are produced through the thermal conversion of the acrylic fibers, for better understanding of the end product (carbon fiber) properties, thermal and oxidative treatments impart an important role (Rahaman, Ismail, & Mustafa, 2007).
Instead of producing carbon fibers out of homo polymer, using a copolymers of polyacrylonitrile improves the fiber quality (Chen & Harrison, 2002). All PAN homo polymer and co-polymers were defined as a PAN-based precursor for carbon fiber mill.
3 1.2.2 PAN-based carbon fibers
As shown in the Figure 1.1 PAN-based carbon fibers have two main steps; stabilization and carbonization and further step was called as graphitization, while converting the PAN-based fibers to carbon fiber. Stabilization step was occurred under oxygen environment and carbonization step takes place under inert atmosphere (nitrogen or argon). There is lots of different parameters effect the strength of PAN precursors, such as interior conditions for instance chemical structure, molecular weight and its distribution, crystallinity and orientation, defects, and exterior conditions such as temperature, tensile speed, and humidity (Wang, Wang, Wu, & Jing, 2007).
4
Schematic view of oxidation step was described in Figure 1.2 by (Rahaman et al., 2007).
Figure 1.2: Schematic view of oxidation.
During stabilization, cyclization and dehydrogenation reactions were occurred which makes PAN fibers denser and stable to keep its fibrous structure for following high temperature carbonization (Edie, 1998; Rahaman et al., 2007).
Carbonization step can be dived into two; low temperature and high temperature carbonization. Carbonization temperature has a critical role on the mechanical properties of the material. Increased temperature assists to increase the carbon content of the material and due to this, high temperature (1000-1500 °C) carbonization the tensile strength increases, according to researches the maximum strength is achieved at 1500 and 1700 oC where the elastic modulus increases monotonically with temperature due to the increased volume fraction and crystallite size, especially at temperatures between 2000 and 3000 oC (Arshad, Naraghi, & Chasiotis, 2011; Edie, 1998; F. Liu, Wang, Xue, Fan, & Zhu, 2008).
5 1.2.3 Electrospinning of nanofiber
It is not possible to produce nanofibers with conventional fiber spinning techniques (wet spinning, dry spinning, melt spinning and gel spinning) generally produce polymer fibers with diameters down to the micrometer range with conventional methods. If the fiber diameter is reduced from micrometers to nanometers, aspect ratio is increased and surface functionalities and better mechanical performance may be achieved (Brown, Stevens, & Stevens, 2007, pp. 71–77).
Spinning of synthetic filaments with the help of electrostatic force is a well-known process for more than a hundred years. First patents using the electrostatic force to produce filaments go to the 1930s by Formhals (Huang, Zhang, Kotaki, & Ramakrishna, 2003)
6
Figure 1.3: Schematic representation of nanofiber production with stable and rotating collector.
Electrospinning parameters can be mainly divided in three main group; Solution parameters, ambient parameters, process parameter and all of these parameters have different effects on the fabricated sample.
1.2.4 Carbon nanofibers
Nowadays, nano-scaled carbon materials such as carbon nanotubes, carbon nanofibers, carbon nanowires etc. have an immense attention because of their unique parameters.
Carbon nanofibers can directly formed via vapor growth or plasma enhanced chemical vapor depositing techniques but these methods are quite complex and high cost processes. In addition, it is possible to be produce carbon fibers out of polymeric nanofiber webs by stabilizing, carbonizing (Gu, Ren, & Wu, 2005).
Electrospinning is a simple and efficient technique for the fabrication of nano to micro scale fibers (Ali & El-Hamid, 2006).
7 2. MATERIALS AND METHODS
2.1 Materials
Polyacrylonitrile was purchased from Sigma Aldrich standard Mw 150,000 g/mol and
was used as received. Dimethylformamide (DMF) was purchased from Sigma Aldrich and was used without any further purification. Acrylonitrile (AN), (99.5> %) and Ammonium persulfate (APS), (99.5> %) were obtained the Aksa Acrylic Chemistry Company and were used as received. Itaconic acid (99.5> %) and Acrylic acid was purchased from Sigma Aldrich. Ethanol and Methanol were all Merck reagents and were used without any purification.
2.2 Equipments and Analysis Scanning electron microscope (SEM)
Scanning electron microscope, is the most common tool for observing the surface morphology and generally used under vacuum conditions and vacuum level is about 10-8 Torr, the accelareted voltage between 1-30 kV, applied during the scanning and electrons are sent through the specimen via electron gun or beam and specimen surface is scanned while beam passing through the optic axis in the meantime detector also shows the secondary electrons or other signals which are backscattered electrons or X-rays and result of scanning across the sample recording and monitors in a two-dimensional raster (Lindsay, 2009, pp. 72–76).
In this study, SEM (Jeol Quanta 200 F FE-SEM) device was used to examine the nanofiber surfaces and diameters of the spunfibers. The acceleration voltage of the SEM was set to 20 kV and samples were coated with thin gold/platinum alloy film using a sputter coater (Balzers Union SKD 030) to prevent the accumulation of charge on their surface.
8
Fourier transform infrared spectroscopy (FTIR-ATR)
Frourier Transform Infrared Spectroscopy instrument is widely used to analyze the structure of organic materials. Main principle of infrared spectroscopy is, measure the infrared absorption by molecular vibrations, result of the molecular vibration infrared spectroscopy shows the wavelength values and generally, wavelength varies between 200 to 4000 cm-1 also crystalline solids make a lattice vibrations which is presented in the range of 20 to 300 cm-1 (Leng, 2010, pp. 235–297).
Fourier analysis (FTIR) presents values according to the IR spectroscopy in signal-to-noise (S/N) ratio, accuracy of the frequency scale, energy throughout, and a ability for versatile data handling (Carbon Materials for Catalysis (eBook), n.d.)
Dynamic mechanical analysis (DMA)
DMA is an instrument, which can apply oscillating force to the sample and according to this force; it examines the material’s response (Menard, 2002). Stress-strain curves and storage and loss modulus curves were obtained using DMA. According to these graphs, it is possible to explain material properties such as stiffness brittleness. In this study, DMA-measurements were performed using a model Q800 from TA instruments.
Differential scanning calorimetry (DSC)
The Modulated Temperature Differential Scanning Calorimetry (MTDSC) measurements were performed using a Q2920 from TA instruments. DSC is the most well known thermal analysis for measuring the thermal properties of the sample such as; solid phase transformation, glass transition, crystallization, melting, and these measurements based on the difference between the sample material and the reference material (Leng, 2010, p. 305). DSC was used for examining the nanofiber webs behaviors under known temperature and examine the duration of the oxidation time for that temperatures. Thermal behavior comparisons between the commercial PAN polymer and synthesized PAN co-polymers was made using DSC and it was used under nitrogen and air environments.
Thermal gravimetric analyze (TGA)
TGA Q 50 from TA instruments was used to investigate the weight change of the sample materials according to the temperature change. Mass loss during the
9
oxidation for different temperatures was also investigated. Under controlled atmosphere, mass change of certain amount of polymer sample according to temperature and oxidation time can be monitored (Menczel & Prime, 2009, pp. 241– 244).
Viscometer
For measuring, the viscosities of the solutions Brookfield LV DV-II+Pro Model Rotational viscometer have been used at room temperature.
Relative humidity and temperature measurement
The room temperature and humidity throughout the electrospinning process, was monitored by Vaisala HMI41 model temperature and humidity measurement device.
11 3. EXPERIMENTAL STUDIES
3.1 Synthesis of Copolymers
Copolymerization of acrylonitrile with acrylic acid and itaconic acid carried in aqueous solution. Reaction medium contains deionized water and DMF. All polymerization experiments were carried out in 3 necked flask and bath temperature was set to 60 oC ± 2 oC. The poly (acrylonitrile-co-acrylic acid) and poly(acrylonitrile-co-itaconic acid) were synthesized by free radical polymerization and APS was used as an initiator in the aqueous medium. First solvent mixture was stirred at 60 0C during 30min, DMF:water mixture volume ratio was 50:50 and kept constant for all reactions. Distilled water was used for the reaction. Then, monomers (AN/AA or AN/IA) were added by dropwise and were stirred during 30 minutes. On the other hand, initiator was dissolved in 20ml of DMF:water mixture by using ultrasonic bath. Finally, initiator was added by dropwise and reaction was occured during 3h. The excessive monomers of acrylic acid, itaconic acid and acrylonitrile were removed by thoroughly washing with ethanol and 3 times with distilled water and polymer was precipitated. The copolymers were then dried in a vacuum oven at 60 oC during 48h. Detailed information about the reaction parameters given in the Table 3.1.
Table 3.1: Polymerization parameters.
Acrylonitrile [%]
Acrylic Acid[%] Itaconic Acid [%] Polymerization medium DMF: Water, 50:50 [ml] Initiator (APS) [g] 99.5 0.5 - 60 0.1 98.0 2.0 - 60 0.1 95.0 5.0 - 60 0.1 93.0 7.0 - 60 0.1 90.0 10.0 - 60 0.2 99.5 - 0.5 60 0.1
12
Table 3.2 (cont.): Polymerization parameters.
98.0 - 2.0 60 0.1
95.0 - 5.0 60 0.1
93.0 - 7.0 60 0.2
3.2 Electrospinning Process
For electrospinning process, Kd Scientific Syringe Series 100 model pump with a working range of 0.1ml/h to 99.9ml/h and Glassman High Voltage Series EH model voltage supply with range in 0 to 30 kV were used for obtaining the nanofiber webs. 20ml Norm-jet van Henke Sass Wolf syringes with 20ml volume and nozzle with inner diameter 1.024mm were used and were purchased from Sigma Aldrich. The rotating wheel was turned with 1680-rpm, collector and flat collector were used for getting non-aligned and aligned nanofiber webs. Usage of rotating collector, is gave an orientation to fiber bundles (Zhou et al., 2009).
3.2.1 Solvent preparation
In this study, commercial homo polymer of polyacrylonitrile and synthesized copolymers of polyacrylonitrile were dissolved in DMF at different weight ratios Table 3.3 and were stirred with magnetic stirrer at room temperature until getting the homogenous solution. These solutions were used for producing the electrospun nanofibers.
13
Table 3.3: Electrospinning parameters.
Weight ratio of PAN in DMF Aligned/Flat (non-aligned) Viscosity mPas Voltage kV Flow Rate ml/h Distance to collector cm Relative Humidity % Temperature ° C 6 F - 10.0 1.2 10 37.8 21.9 7 A 160 17.0 1.2 10 28.5 21.5 7 F 160 12.0 1.2 10 31.3 19.0 8 A 290.3 17.5 1.2 10 27.8 22.6 8 F 290.3 12.0 1.2 10 28.5 22.8 9 A 594.5 16.5 1.2 10 21.6 22.0 9 F 594.5 10.0 1.2 10 28.8 21.4 10 A 823 16.5 1.2 10 28.2 22.9 10 F 823 15.0 1.2 10 29.3 22.3 11 A 1405 13.5 1.2 10 28.4 22.0 11 F 1405 13.0 1.2 10 28.8 21.4
14 3.2.2 Sample preparation
Electrospinning parameters for each sample given in the Table 3.3 Samples were spinned both aligned and flat formation using different process parameters for obtaining good quality nanofiber webs. Aligned nanofiber webs were obtained by using the rotating collector and for flat nanofiber webs were spunned on the fixed collector.
3.3 Oxidation
Thermal oxidation of polyacrylonitrile and copolymers of polyacrylonitrile at temperatures between 200-300 oC has been studied. PAN based nanofiber webs used as a precursor for this step and produced by the heating of PAN based nanofibers at various temperatures inside the furnace at air atmosphere within this range for different periods, up to 3h. Stretching was applied during the all oxidation experiments. During the stabilization process, for preventing the shrinkage and maintain molecular orientation, tension should be applied to the sample and thermal oxidation at 200-300 oC provide cyclization and the formation of a thermally stable aromatic ladder polymer (Chen & Harrison, 2002).
3.4 Carbonization
Final step of the producing carbon nanofibers is called as carbonization. Removing the other elements from the non-woven web was aimed for carbonization step. For carbonization step, low temperature carbonization applied at 950 oC during 1h with stretching under nitrogen environment. Heating time was took app. 4.5h and after 1h waiting time for carbonization samples were not taken immediately from the furnace for preventing the sudden damages on the sample. After waiting 1 more hour for cooling down Nitrogen gases were turned off and sample was taken from the inside of the furnace.
15 4. RESULTS AND DISCUSSIONS
4.1. Characterization of Polymers and Nanofibers 4.1.1. FTIR-ATR spectroscopy
The FTIR-ATR spectra of AN-IA and AN-AA copolymers for different feed ratios are shown in Figure and Figure respectively and it was recorded in the absorbance mode.
In Figure 4. the characteristic peak of C≡N stretching of AN repeating unit is observed as a strong absorption peak at around 2244 cm-1. The peak at about 1735 and 1628 cm-1 are related to the carbonyl stretching vibration of the carboxylic acid. The peak of carbonyl stretching in copolymers is shifted from 1737 to 1732 cm-1, corresponding to increase in IA content. The characteristic peak of aliphatic -CH2-
stretching is at 2940 cm-1. There is also a strong band at 1454 related to bending vibration of -CH in -CH2. The band at 1076 cm-1 is ascribed to the -CH bending
mode in CH and also characteristic peak C=O which was observed due to addition of the monomer units, was observed between 1735-1710 cm-1. These results are in agreement with (Bhanu et al., 2002; Devasia, Nair, Sivadasan, Katherine, & Ninan, 2003; Moghadam & Bahrami, 2005; Wangxi, Jie, & Gang, 2003).
16
Figure 4.1: FTIR-ATR spectra of PAN powder, film and nanofiber.
3500 3000 2500 2000 1500 1000 500 0.0 0.1 0.2 0.3 0.4 0.5 Ab so rb an ce Wavelength 0.5 % AA 2% AA 5 AA 7 AA 10 % AA
Figure 4.2: FTIR-ATR spectra of P(AN-AA) co-polymer for different feeding of wt% of AA.
17
Figure 4.3: Correlation between FTIR-ATR absorbance ratio( CN/C=O peaks) with feeding wt% of AA.
Figure 4.4: FTIR-ATR spectra of P(AN-IA), feeding weight ratio of AN/IA, 93/7.
3500 3000 2500 2000 1500 1000 0.00 0.02 0.04 0.06 0.08 0.10 Ab so rb a n ce wavelength P(AN-IA) 93/7 0 2 4 6 8 10 0.2 0.4 0.6 0.8 1.0 1.2 1.4 P(AN-AA) AA content C=O/CN AA co n te n t (R a tio 1 7 3 0 /2 2 4 0 ) wt% of AA feeding Equation y = a + Adj. R-Squ 0.7105 Value Standard E Book3_B Intercep 0.441 0.14287 Book3_B Slope 0.078 0.02393
18
Figure 4.5: FTIR-ATR spectra of P(AN-AA), feeding weight ratio of AN/AA, 93/7. Viscosity vs. concentration graph was come up as a polynomial function, which is increasing, with the increased polymer amount inside the solvent, as expected.
6 7 8 9 10 11 12 0 500 1000 1500 2000 2500 Vis cos ity [ m Pas ] wt % of PAN
Figure 4.6: Weight ratio of PAN polymer inside the solvent vs. viscosity.
3500 3000 2500 2000 1500 1000 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 Ab so rb a n ce wavelength P(AN-AA) 93/7
19 4.1.2 Morphology of electrospun nanofibers
Fiber diameter measurement was based on the SEM images; minimum measured sample number was 50 and for determining the diameter of the sample ImageJ software was used.
Aligned nanofiber webs have thinner diameter compared to flat group and for lower concentration of PAN such as 6% and 7%, beat formation was observed on the electrospun web. Also, standard deviation of these webs are higher than the others. In addition, standard deviation of the fiber diameters is higher for low PAN concentration.
20
Figure 4.7: SEM images of electrospun flat nanofiber webs; (a) 6wt% PAN, (b) 7wt% PAN, (c) 8wt% PAN - scale is 2 micrometer and (d) 9wt% PAN, (e) 10wt% PAN, (f) 11wt% PAN and scale is 5 micrometer.
21
Figure 4.8: Average diameter distribution vs. wt% of PAN.
100 200 300 400 500 0 5 10 15 20 # o f f ib e rs diameter [nm] Average diameter 164.81 nm SD 79.80 CV% 48.42 Figure 4.9: Diameter histogram for 6 wt% PAN flat nanofiber web.
22
Figure 4.10: SEM images of electrospun aligned nanofiber webs; (a) 6wt% PAN , (b) 7wt% PAN, (c) 8wt% PAN d (d) 9wt% PAN, (e) 10wt% PAN, (f) 11wt% PAN (scale is 2,1,1,2,2,5 micrometer, respectively).
23 4.1.3 Differential scanning calorimetry (DSC)
As seen from the DSC results co-polymers have a positive effect to shift to peak point to the lower temperature level, compared to homo polymer PAN. When DSC curves of nanofiber web and film were compared, there is not much difference between the peak of the fiber web and film (Wang-xi Zhang, Wang, & Sun, 2007) it was also shown in Figure. Co-polymers of PAN can be show a doublet exothermic peak on DSC measurements (Ouyang, Cheng, Wang, & Li, 2008; Wangxi Zhang & Li, 2005; Wang-xi Zhang et al., 2007).
Figure 4.11: DSC curves for P(AN-IA), P(AN-AA) co-polymers and homo polymer PAN 0 100 200 300 400 -1 0 1 2 3 4 5 299 o C 270 o C H e a t F lo w W /g Temperature oC PANIA937 PANAA937 PAN 250 oC
24
Figure 4.12: DSC compration between powder formation and nanofiber formation of PAN.
It was marked that the DSC curves obtained from the co-polymer, electrospun nanofiber had two exothermic peaks instead of one exothermic peak in homopolyacrylonitrile and this can be because of, two types of thermo-chemical reactions occurred in the electrospun nanofibers (J. Liu et al., 2009).
0 100 200 300 400 0 2 4 6 8 10 332 oC H ea t F lo w W /g Temperature o C 8% P(AN-AA) 93/7 ANF 277 oC
Figure 4.13: DSC curve for 8%wt of P(AN-AA) co-polymer, aligned nanofiber formation. 0 100 200 300 400 -2 -1 0 1 2 3 4 5 H e a t F lo e W /g temperature oC Nanofiber Powder
25 4.1.4 Thermal gravimetric analysis (TGA)
While temperature increases through the 100 °C, moisture content of the sample was eliminated and after that temperature, real mass loss of the material can be observed.
Figure 4.14: TGA comparison of PAN nanofiber web, film and powder.
4.1.5 Dynamic mechanical analysis (DMA)
Tg values of nanofiber webs were measured by using DMA multi frequency strain mode. As shown in the Figure 4.17 and Figure 4.18, Tg of aligned PAN based nanofiber was measured as 99.66 oC and P (AN-AA) based nanofiber was measured as 95.67 oC.
When sample was in aligned formation, strain value of the sample decreases and strength of the sample increases. Because, oriented fibers have already aligned and these oriented fibers help to improve strength. All stretching procedures have a positive effect on the sample to increase its mechanical properties, it can be stretching process on the sample or production technique which can give an opportunity to produce oriented fiber web (Hou, Yang, Zhang, Waclawik, & Wu, 2010). Detailed DMA curves for aligned and non-aligned nanofiber webs were presented in appendices part.
0 100 200 300 400 500 600 0 20 40 60 80 100 w e ig h t l o st % Temperature oC PAN Film PAN Powder PAN NF
26
Figure 4.15: Stress-strain curve for aligned nanofiber web with average value and multiple form.
Figure 4.16: Stress-strain curve for flat nanofiber web with average value and multiple form.
27 20 40 60 80 100 120 140 160 180 200 0.0 0.2 0.4 0.6 0.8 1.0 T a n D e lta Temperature oC P(AN-AA) 93/7 ANF Tg 95.67 oC
Figure 4.17: Tan delta curve from DMA measurement for investigating Tg value for P(AN-AA), 93/7 co-polymer, aligned nanofiber web with error bars.
40 60 80 100 120 140 160 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 PAN ANF Tg 99.66 oC T a n D e lta Temperature oC
Figure 4.18: Tandelta curve from DMA measurement for investigating Tg value for PAN homo polymer, aligned nanofiber web with error bars.
28 20 40 60 80 100 120 140 160 0.2 0.4 0.6 0.8 1.0 T a n D e lta Temperature oC PAN P(AN-AA)
Figure 4.19: Comparison of PAN and P(AN-AA) aligned nanofiber tandelta curves for Tg values
Figure 4.20: Correlation between Young’s Modulus and wt% of PAN both for aligned and flat electrospun fiber webs.
7.0 7.5 8.0 8.5 9.0 9.5 10.0 400 600 800 1000 1200 1400 1600 1800 2000 2200 Flat Aligned Young's M odulus M Pa wt % of PAN
29 0 200 400 600 800 1000 1200 1400 1600 100 200 300 400 500 600 700 800 900 1000 flat aligned Avr. fi be r di ame te r [n m] Viscosity [mPas]
Figure 4.21: Relationship between average fiber diameter and viscosity both for aligned and flat electrospun fibers.
Figure 4.22: Combination of Young’s Modulus and average fiber diameters.
150 200 250 300 350 400 450 500 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Aligned Flat Yo u n g 's Mo d u lu s MPa
30
4.2 Characterization of Oxidized and Carbonized Nanofibers 4.2.1 FTIR-ATR spectroscopy
For different oxidation times and temperatures FTIR-ATR curves measured and characteristic peak of polyacrylonitrile about 2244 cm-1 decreased but never disappeared end of the oxidation step. For better understanding of oxidation degree C≡N/C=O ratio calculated and decreased ratio represents the enhanced oxidation level. During the oxidation step nitrogen bonding between carbon(C≡N) never was gone but was changed triple bond formation to double bond because of that, during the oxidation stage, intensity of the C≡N bond was decreased while intensity of C=O bond was increased (Dalton, Heatley, & Budd, 1999; J. Liu et al., 2009; Ogawa & Saito, 1995; Standage & Matkowsky, 1971).
Figure 4.23: FTIR-ATR spectra of oxidized PAN aligned nanofiber webs at 235 oC for different durations.
3500 3000 2500 2000 1500 1000 500 0.00 0.02 0.04 0.06 0.08 0.10 0.12 Ab so rb a n ce wavelength 3h 2h 30min Oxidation @ 235 oc
31
Figure 4.24: FTIR-ATR spectra of oxidized PAN aligned nanofiber webs at 270 oC for different durations.
230 240 250 260 270 280 290 300 310 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 3h oxidation Abs orbanc e R at io C N /C =O Temperature
Figure 4.25: Correlation between FTIR-ATR absorbance ratio of CN/C=O with temperature. 3500 3000 2500 2000 1500 1000 500 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Ab so rb a n ce wavelength 3h 1h 30min Oxidation @ 270 oC
32
Figure 4.26: FTIR-ATR spectra of oxidized PAN aligned nanofiber webs for different oxidation temperatures, oxidation duration 3h
Figure 4.27: FTIR-ATR spectra comparison between oxidation at 235 and 270
o C during 30 min. 3500 3000 2500 2000 1500 1000 500 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 Ab so rb a n ce wavelength 300C 270C 250C 235C oxidation time 3h 3500 3000 2500 2000 1500 1000 500 0.00 0.05 0.10 0.15 0.20 0.25 0.30 a b so rb a n ce wavelength 270C 235C oxidation time 30 min
33 20 40 60 80 100 120 140 160 180 200 0.0 0.2 0.4 0.6 0.8 1.0 Abs orbanc e R at io C N /C =O @235C oxidation different time periods
time (min)
Figure 4.28: Correlation between FTIR-ATR absorbance ratio for different oxidation duration, oxidation temperature 235 oC.
20 40 60 80 100 120 140 160 180 200 0.0 0.2 0.4 0.6 0.8 1.0 @270 C oxidation different time periods
Abs orbanc e R at io C N /C =O time (min)
Figure 4.29: Correlation between FTIR-ATR absorbance ratio for different oxidation durations, oxidation temperature 270 oC.
34 20 40 60 80 100 120 140 160 180 200 0.0 0.2 0.4 0.6 0.8 1.0 235 270 time (min) 0.02 0.03 0.04 0.05 0.06 0.07
Figure 4.30: Correlation between FTIR-ATR absorbance ratio for different oxidation durations, oxidation temperatures 235 and 270 oC.
20 40 60 80 100 120 140 160 180 200 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 P(AN-AA) after oxidation @270 oC C N /C =O duraiton of oxidation
Figure 4.31: Relationship between duration of oxidation and FTIR-ATR absorbance ratio for P(AN-AA) nanofiber web.
35 4000 3500 3000 2500 2000 1500 1000 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22 0.24 0.26 Ab so rb a n ce wavelength w/o oxidation 30min 2h 3h
Figure 4.32: FTIR-ATR spectra of P (AN-AA), oxidation @ 270 oC for different time periods.
4.2.2 Morphology of nanofibers
After oxidation, white color of nanofiber webs turned to yellow, tan brown dark brown after oxidation step, increased time and temperature increased the intensity of the color, and finally after carbonization it was appeared as a black. Fiber diameter was shifted to lower values after oxidation and carbonization steps due to the mass loss.
The result of heating of the PAN nanofiber webs under oxygen environment, the resulting ladder structures formed in the polymers caused the initial changes in color of nanofibers webs (Wang-xi Zhang et al., 2007) and this color change can be seen in the Figure. Carbon content of the fiber increases through the carbonization steps. Homo polymer PAN include ~68% carbon and 26% nitrogen after stabilization it reaches 65% and nitrogen content decreases to22% end of low heat temperature carbonization, more than 92% carbon estimated inside the fiber (Fitzer, 1989).
36
Figure 4.33: Color change after oxidation and carbonization from left to right; oxidation at 235 °C-30 min.,oxidation 235 2h, oxidation 300 °C-1h, Carbonization at 950 °C-1h.
Aligned nanofiber web, which is produced 10% of PAN, was selected as a precursor for oxidation and carbonization steps. Fiber diameter distribution for selected web is shown in the Figure with SEM image of the same web. It can be easily seen from the fiber diameter histograms, fiber diameter through the steps was decreased because of the applied high temperature and weight loss on the sample.
Figure 4.34: 10% PAN ANF as precursor, histogram of fiber diameter and SEM image.
37
For lower concentrations fiber diameter varies a lot thus, standard deviation of the fiber diameter value was higher also beat formation was shown and irregularities during the fiber was observed and these results was given in the appendices part.
Figure 4.35: 10% PAN ANF was used as precursor and (a) after oxidation at 270 C-1h, (b) after carbonization of (b), (c) after oxidation at 270 C-3h, (d) after carbonization of (c).
38 200 300 400 500 600 700 800 0 2 4 6 8 10 12 14 16 18 20 22 # o f f ib e rs Diameter [nm] ANF oxi 270 3h CNF
Figure 4.36: Fiber diameter change on sample 10% PAN ANF, after oxidation at 270 C- 3h and carbonization at 950 C- 1h. 200 300 400 500 600 700 800 0 2 4 6 8 10 12 14 16 18 Diameter [nm] # o f f ib e rs ANF Oxi @2700-1h CNF
Figure 4.37: Fiber diameter change on sample 10% PAN ANF, after oxidation at 270 C- 1h and carbonization at 950 C- 1h.
39
Figure 4.38: (a) 8% P(AN-AA) ANF, (b) after oxidation at 270 C-3h, (c) after carbonization at 950 C-1h.
Figure 4.39: Fiber diameter change on sample 8% P(AN-AA) ANF, after oxidation at 270 C- 3h and carbonization at 950 C- 1h.
4.2.3 Differential scanning calorimetry (DSC)
Time and temperature are the two main parameters that affect the efficiency of the oxidation. As shown in the figure, fractional conversion at certain time plotted for different temperatures and the reaction rate is faster.
100 150 200 250 300 350 0 2 4 6 8 10 12 14 16 18 20 Diameter [nm] # o f f ib e rs oxi NF CNF ANF
40 0 200 400 600 800 1000 0.0 0.2 0.4 0.6 0.8 1.0 F ra ct io n a l C o n ve rsi o n ( ) time (min) 235 250 270 300
Figure 4.40: Conversion-time plots at different temperatures.
The conversion (α) at any time (t) was obtained from the relation 1 where ∆Hf is the fractional heat of reaction and ∆H is the total enthalpy (1).
Table 4.4: DSC-enthalpy values for different oxidation duration and temperature. Oxidation Temperature-Time (°C-min) Isothermal Temp-Time (°C-min) Enthalpy J/g Isothermal Temp-Time (°C-min) Enthalpy J/g 235 °C- 180 500 235 2358.0 500 270 1706.0 270 °C- 180 500 270 840.5 500 300 414.0 300 °C- 180 500 300 310.1 - - -
4.2.4 Dynamic mechanical analysis (DMA)
It was seen that, after oxidation step there is not significant effect of the polymer amount inside the electrospinning solvent on the Young’s modulus value of the web. Stretching has a positive effect on the fiber web for improving the mechanical properties (Cai, Chen, Nesterenko, & Meyers, 2008; Lai et al., 2011; Qin, Lu, Xiao, Hao, & Pan, 2011).
41
Table 4.5: Average Young’s Modulus values after oxidation. Sample content Oxidation
Temperature o C Oxidation time min Young’s Modulus MPa 8% PAN ANF 235 30 3610.25 8% PAN ANF 235 120 3396.57 8% PAN ANF 235 180 4089.75 9% PAN ANF 250 180 3499.75 8% PAN ANF 270 180 3911.00 11% PAN ANF 300 180 2623.67 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 0 10 20 30 40 50 60 70
8% PAN ANF oxi @235 C-2h avr. Young's Modulus 3396.57 MPa St re ss [MPa ] Strain %
42 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 10 20 30 40 50 60 70 80 St re ss [MPa ]
9% PAN ANF oxi @250 C-3h avr. Young's Modulus 3499.75 MPa
Strain %
Figure 4.42: DMA stress-strain curve for 9% PAN ANF, oxidation at 250 oC-3h.
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0 10 20 30 40 50 60 70 80 St re ss [MPa ]
8% PAN ANF oxi @235 C-30 min avr. Young's Modulus
3610.25 MPa
Strain %
43 0.0 0.1 0.2 0.3 0.4 0.5 0 1 2 3 4 5 6 Strain % St re ss [MPa ] 8 % P(AN-AA) 93/7 CNF Young's Modulus 1509 MPa
Figure 4.44: DMA stress-strain curve for 8% P(AN-AA) ANF, oxidation at 270 o C-2h. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 10 20 30 40 50 60 70 80 St re ss [MPa ]
8% PAN ANF oxi @235 C-3h avr. Young's Modulus 4089.75 MPa
Strain %
44 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 5 10 15 20 25 30 35 40 45 50 St re ss [MPa ]
8% PAN ANF oxi @270 C-3h avr. Young's Modulus 3911 MPa
Strain %
Figure 4.46: DMA stress-strain curve for 8% PAN ANF, oxidation at 270 oC-3h.
Figure 4.47: DMA stress-strain curve for 11% PAN ANF, oxidation at 270 oC-2h and carbonization at 950 oC-1h. 0.0 0.2 0.4 0.6 0.8 1.0 0 10 20 30 40 50 60 70 11% PAN CNF oxi @270 C-3h carbonization @950C-1h avr. Young's Modulus 10881.2 MPa St re ss [MPa ] Strain %
45 0.0 0.1 0.2 0.3 0.4 0.5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 10% PAN CNF oxi @270 C-3h carbonization @950C-1h avr. Young's Modulus 10211.75 MPa St re ss [MPa ] Strain %
Figure 4.48: DMA stress-strain curve for 10% PAN ANF, oxidation at 270 oC-3h and carbonization at 950 oC-1h. 0.0 0.2 0.4 0.6 0.8 1.0 0 10 20 30 40 50 60 70 10% PAN CNF oxi @270 C-1h carbonization @950C-1h avr. Young's Modulus 9560 MPa St re ss [MPa ] Strain %
Figure 4.49: DMA stress-strain curve for 10% PAN ANF, oxidation at 270 oC-1h and carbonization at 950 oC-1h.
46 4.2.5 Thermal gravimetric analysis (TGA)
For better understanding of mass loss during oxidation step for different temperatures, TGA measurements were done for PAN-based nanofiber webs. After isothermal at certain time period at 200-300 oC for oxidation and 950 oC for carbonization weight loss was occurred. Thus, this weight loss affects the fiber diameter. Fibers are in cylindrical form and weight loss was occurred on the surface. Because of this, core shell structure can be existed and it has an irregularity effect on the sample which decreases the strength (J. Liu et al., 2009).
60 80 100 120 140 160 180 200 220 240 260 280 300 85 90 95 100 w ei gh t % Temperature oC 235 C 250 C 270 C 300 C
Figure 4.50: TGA curve for same isothermal time (300min) for different temperatues.
Figure 4.51: Relationship between weight loss and temperature. 30 28 26 24 22 20 18 16 14 230 240 250 260 270 280 290 300 310 T emp era tu re weight loss %
47
It is seen in the Figure 4.51 percentage of weight loss increased through the increased temperature as expected. If you give more heat to the sample then chemical reactions speed and amount increases and it was end up with more weight loss on the sample.
Table 4.6: % of weight loss after waiting 300 min. at different temperatures. Weight Loss % Temperature o C Isothermal min 14.90 235 300 20.58 250 300 24.4 270 300 29.75 300 300
49
5. CONCLUSIONS AND RECOMMENDATIONS
In this study, carbon nanofibers out of PAN precursor have been successfully produced. Homopolymer of polyacrylonitrile and synthesized co-polymers of acrylonitrile, was used as a PAN-based precursor of carbon nanofiber production line. Production of carbon nanofiber process is a time-consuming process and energy-consuming level is high because of the high temperature parameters. Synthesized co-polymers decreased the Tg value thanks to this property lower oxidation temperature can be used and it helps to consume less energy during the process. Addition of monomers while synthesizing polyacrylonitrile co-polymers was created positive effects on the sample such as decreased on the Tg value. It means that, initiation temperature for oxiditaion process was also increased due to lower Tg value and it is easier to control the process at lower temperatures.
Aligned nanofiber webs have a higher strength value compared to non-aligned webs. In this perspective, usage of a rotating wheel as an electrospinning collector was suggested to use while producing carbon nanofiber precursor out of PAN solutions. An effect of oxidation on the sample was not seen in terms of strength and young’s modulus. After oxidation sample was became more frigle. However, FTIR-ATR curves show that, increased oxidation time and temperature increase the oxidation level.
For further studies instead of low temperature carbonization, high temperature carbonization can be also applied. It will improve the strength of the sample. Electrospinning of co-polymers can be studies harder for optimization of the parameters because precursor of PAN-based nanofiber webs directly affects the end product quality. Also, some solutions for handling problem of carbon nanofiber webs should be investigated to get better results from produced carbon nanofibers
51 REFERENCES
Ali, A. A., & El-Hamid, M. A. (2006). Electro-spinning optimization for precursor carbon nanofibers. Composites Part A: Applied Science and Manufacturing, 37(10), 1681–1687.
Arshad, S. N., Naraghi, M., & Chasiotis, I. (2011). Strong carbon nanofibers from electrospun polyacrylonitrile. Carbon, 49(5), 1710–1719.
Bashir, Z. (1991). A critical review of the stabilisation of polyacrylonitrile. Carbon, 29(8), 1081–1090. doi:10.1016/0008-6223(91)90024-D
Bhanu, V. ., Rangarajan, P., Wiles, K., Bortner, M., Sankarpandian, M., Godshall, D., McGrath, J. . (2002). Synthesis and characterization of acrylonitrile methyl acrylate statistical copolymers as melt processable carbon fiber precursors. Polymer, 43(18), 4841–4850.
Brown, P. J., Stevens, K., & Stevens, K. A. (2007). Nanofibers and nanotechnology in textiles. Woodhead.
Cai, J., Chen, Y., Nesterenko, V. F., & Meyers, M. A. (2008). Effect of strain rate on the compressive mechanical properties of aluminum alloy matrix
composite filled with discontinuous carbon fibers. Materials Science and Engineering: A, 485(1–2), 681–689.
Chen, J. ., & Harrison, I. . (2002). Modification of polyacrylonitrile (PAN) carbon fiber precursor via post-spinning plasticization and stretching in dimethyl formamide (DMF). Carbon, 40(1), 25–45.
Dalton, S., Heatley, F., & Budd, P. M. (1999). Thermal stabilization of polyacrylonitrile fibres. Polymer, 40(20), 5531–5543.
Devasia, R., Nair, C. P. R., Sivadasan, P., Katherine, B. K., & Ninan, K. N. (2003). Cyclization reaction in poly(acrylonitrile/itaconic acid) copolymer: An isothermal differential scanning calorimetry kinetic study. Journal of Applied Polymer Science, 88(4), 915–920.
Edie, D. D. (1998). The effect of processing on the structure and properties of carbon fibers. Carbon, 36(4), 345–362.
Fitzer, E. (1989). Pan-based carbon fibers—present state and trend of the technology from the viewpoint of possibilities and limits to influence and to control the fiber properties by the process parameters. Carbon, 27(5), 621–645. Gu, S. Y., Ren, J., & Wu, Q. L. (2005). Preparation and structures of electrospun
PAN nanofibers as a precursor of carbon nanofibers. Synthetic Metals, 155(1), 157–161.
