New Conversion of Solar Energy to Electricity by Natural Dye Sensitization 1 2 for merge1 2

Conversion of Solar Energy to Electricity by NaturalDye-Sensitization

C.I.F.Attanayake 1 B.A.J.K.Premachandra 1 A. A.P. de Alwis 1. andG.K.R.Senadheera 2.

1.Department of Chemical and Process Engineering, University of Moratuwa, Moratuwa, Sri Lanka

2.Institute of Fundamental Studies, Hantana Road, Kandy , Open University of Sri Lanka , Kandy Regional Centre, Kandy, Sri Lanka.

Corresponding Author , e-mail : [email protected]

Abstract

Preliminary investigations on the identification of naturalpigments in the dye-sensitization of nanocrystalline n-typeTiO2 were carried out. Fresh extracts of Mangoostein, Rambutan,Mango, Tomato, Carrot, King Coconut, Pumpkin, Red Banana,Beetroot, Turmeric, Venivel, Orange, Grape, Spinach, Wattakka,Ginger etc were employed as sensitizers in thin layer sandwichtype photo electrochemical dye – sensitized solar cells(DSSC's).

After electrical and electronic analysis, it was observed thatmany useful dyes which could be extracted from naturalproducts by simple procedure could be used as photosensitizers for DSSC's. It was also observed that dye extractsof Turmeric and Mangoostein yielded better results.

The current-voltage curves obtained with solar cells employingthe photo anode with TiO2 sensitized by different dyes wereobserved. The values of short circuit current density (Jsc),open circuit voltage (Voc), fill factor (ff), and efficiency(η) obtained for solar cells employing photo anodes with TiO2

sensitized with different fruit / vegetable extracts werenoted . The dye extracts of Turmeric root and Mangoosteinfruit were found to be superior to those obtained from otherdyes , and were Jsc = 0.540 mAcm-2 and 0.444 mAcm-2, Voc =599.1 mV and 565.2 mV, ff = 69.03 % and 65.66 % , η = 0.223 %and 0.165 % respectively.

This work done incorporates the foundation on which researchcould be done to develop low-cost, high efficiency solar

1

energy to electricity conversion units for use in countrieslike Sri Lanka.

1. Introduction

Technological achievements in the clean energy systeminfrastructure are a fundamental issue for worldwide economyand environmental improvements. Therefore, in this 21st

century, energy based non-renewable sources has to beconverted into new energy systems by incorporating noveltechnologies derived from advancements in science [1]. Amongseveral new energy technologies, Die-Sensitized Solar Cells(DSSC's) are one of the most promising new energy generationsystems for photovoltaic technology. It has emerged as one ofrenewable energy sources as a result of exploiting several newconcepts and materials, such as nanotechnology and moleculardevices. Even though the first dye sensitization ofsemiconductors was reported by Vogel in 1873, where Silverhalide emulsions were sensitized by dyes to produce black andwhite photographic films, the use of dye sensitization inphotovoltaic’s had been achieved little noticeable resultuntil a break through at the early 1990's by Gratzel's group.They developed a DSSC consisting of TiO2 electrode sensitizedwith Ruthenium (II) complex dye, organic liquid electrolytewith iodine/iodide red ox couple and Platinum depositedcounter electrode. The solar energy to electricity conversionefficiencies were reported on high as 7.1% in 1991 and 10% and11% in 2008 [2] . In these devices a monolayer of the dye isdirectly attached to the semiconductor surface via carboxylgroup, which could realize an efficient injection of chargecarriers from photo excited dye to semiconductor. However thissensitization of TiO2 for solar applications requires not onlyefficient but also stable and inexpensive sensitizers. So far,several organic dyes and organic metal complexes have beenemployed to sensitize nanocrystalline TiO2 semiconductors andone of the most efficient sensitizer is transition metalcoordination compound (Ruthenium polypyridyl complex). This isbecause the complex has intense charge – transfer (CT)

2

absorption in the whole visible range, long excited lifetimeand highly efficient metal-to-ligand charge transfer (MLCT).

However, other organic dyes, such as phythalocyanine, cyaninedyes, xanthalene dyes, coumarin dyes and so on usually performpoorly in DSSC's because of weak binding energy with TiO2 filmand low charge transfer absorption in the whole visible range,but these dyes are very cheap and can be prepared easily,compared to Ruthenium polypyridyl complexes. On the otherhand, in nature, the fruit, flower, root and leaf of plantsshow various colours from red to purple and contain variousnatural dyes which can be extracted by simple procedure.Therefore, it has been emphasized by many researchers toobtain useful dyes as photo sensitizers for DSSC's fromnatural products, because of the simple preparationtechniques, widely available sources, and low cost [3],[5].Due to these reasons the importance of work done by theauthors to develop low cost solar energy to electricityconversion units in principle is emphasized.

All research done on DSSC”s in Sri Lanka have been performedat the Institute of Fundamental Studies (IFS) , Hantana, Kandysince 1994 by research teams led by Professor K.Tennakone .Most of the research at the IFS has been done using veryexpensive Ruthanium bipiridyl metallic synthetic dyes and haveachieved the highest conversion efficiency of 10.0 % in May2001 .Very limited research on DSSC”s have been done in SriLanka at the University of Peradeniya , University of Ruhuna ,University of Kelaniya , University of Colombo , University ofJaffna , and Open University of Sri Lanka using varioussynthetic dyes and metallic oxides , but not using naturaldyes of plants .Only the Authors of this paper have done anyserious research on DSSC”s using natural dyes of plants grownin Sri Lanka .

A major disadvantage of DSSC”s is that they degrade whenexposed to ultra violet light . The barrier layer tocounteract this may include UV absorbing luminescent

3

chromospheres (which emits at longer wavelengths ) andantioxidants to protect and impose the efficiency of the cell.

Another major disadvantage is of the DSSC design is the use ofliquid electrolyte which has temperature stability problems .At lower temperatures the electrolyte can freeze ending powerproduction and potentially leading to physical damage . Highertemperatures can cause the liquid to expand , making thesealing of the panels a serious problem . Another drawback isthe electrolyte solution which contains volatile organicsolvents and must be carefully sealed . Replacing the liquidelectrolyte with a solid has been a major ongoing field ofresearch .

2. Methodology

2.1 Extraction of dyes

The extracts of dyes from various fruits and vegetables wereobtained from fresh fruits and vegetables. The clean fruitsand vegetables were crushed and added to Ethanol (Merck). Whennecessary, the mixtures were centrifuged and all solutionswere protected from direct light exposure.

2.2 Preparation of nanocrystalline TiO2 films

A TiO2 paste was prepared by blending 200 mg powder of TiO2 (P-25, Degussa) and 1 drop of Triton X 100 in an agate mortar,then the mixture was ground for 30 min whilst adding 10 dropsof Acetic acid , finally 10ml of Ethanol was slowly addedwhilst grinding continuously for the another 30 min. Abovepastes were then applied on Fluorine-doped Tin oxide coatedtransparent conducting glass substrate by well known DoctorBlade method to obtain approx 10 micrometer thick TiO2 film.Films were then heat treated at 5500C for 30 min. and cooleddown to room temperature. Then they were immersed separatelyin alcoholic dye solutions for 12 hours.

4

2.3 Fabrication and characterization of solar cells

Photo electrochemical solar cells were then fabricated bysandwiching a Platinum sputtered conducting Tin oxide (CTO)glass plate with the dyed TiO2 films. A red ox electrolytecontaining / redox couple was then introduced to the

solar cells. I-V characteristics of the solar cells at 100mWcm-

2 (AM 1.5) were measured using a home-made I-V measuring set upcoupled with Keithley 2000 Electronic Multimeter with aPotentiostat via a computer controlled software available atthe Institute of Fundamental Studies (IFS), Kandy. Xenon 500lamp was also used with AM 1.5 filters to obtain simulatedsunlight with the intensity of 100 mWcm-2. The intensity of thelight was calibrated using an EKO Pyronometer and Siliconphotodiode. The Absorption Spectra were also obtained with theUV 2450 SHIMADZU UV-VIS Spectrophotometer available at theIFS.

2.5 General Error Analysis

The main errors encountered during the characterization ofDSSC”s are the instrumental errors during the measurement ofOpen Circuit Voltage (mV), Short Circuit Current (mA) , FillFactor (%), and Conversion Efficiency (%) by using theKeithley 2000 Electronic Multimeter with Potentiostat coupledto a home –made IV curve tracing Computerized SoftwareSystem . The summation of these errors constitute a totalpossible error of + or – 1 % in the determination of theseparameters and could be considered as acceptable .

3. Results and Discussion

3.1 Absorption Spectra obtained from various fruits andvegetables

Figure 1 depicts the absorption spectra of ethanolic dyesolution. It can be seen that, the dye solution obtained from

5

the Mangoostein absorb more in the red side of the spectrumthan the other dyes.

The current voltage curves obtained with a few solar cellsmeasured with dyes initially employing the photo anode withTiO2 sensitized by different dyes is presented in Figure 2.Table 1 presents the values of short circuit current density(Jsc), open circuit voltage (Voc), fill factor (ff), andefficiency () obtained for solar cells employing photo anodeswith TiO2 sensitized with different fruit / vegetable extracts.Fill factor values from 0.30 to 0.64 were obtained with thesedyes. The average efficiency values obtained for cells havingthe dye from Mangoostein fruit is superior to those obtainedfrom other dyes. This could be due to better interactionbetween the dye molecules and the surface of TiO2.

Subsequent analysis and evaluation of further samples ofnatural dyes extracted from a cross section of plants grown inSri Lanka is presented at Table 2 in Appendix. From this Tableit is observed that extracts of Turmeric rhizome root (darkyellow ) and extracts of Mangoostin fruit rind (dark purple )yielded better results amongst these dyes tested. It has beenobserved from these results that darker the colour of thesedyes, greater would be their conversion efficiencies . Alsothe addition of trace amounts of Acetic acid , Hydrochloricacid etc to these natural dyes have been observed duringrecent tests to possibly change their chemical structures ,and hence to make them appear darker in colour and to increasetheir conversion efficiencies .

The most probable target conversion efficiencies of thesenatural dyes suitable for use in DSSC “s would be around 3%-- 5 % . The Authors of this paper have recently obtained apractical conversion efficiency of 0.5 % for Mangoostin fruitrind , and nearly 1.0 % for Fire Fern leaf (deep purplishbrown ) garden plant endemic to Ecudor , Venezuela andColombia in Central America, recently brought down toPeradeniya Botanical Gardens . Even though the efficiency

6

350 400 450 500 550 600 650 700 750 800W ave length (nm )

Norm

alize

d Ab

sorb

ance

ad hbc

g

i

f

e

values obtained in this study are not significant with thevalues obtained in the system with Ruthenium complexes, thestraight forward preparation of photo anodes withsemiconductor oxides sensitized by natural dyes still enables,a much cheaper and easy environmentally friendly productionof solar cells. Further it provides an interesting and cheapalternative to commonly used very expensive and raresynthetic dyes. Therefore, investigations are being carriedout in searching for efficient natural dyes which can havepotential use in these DSSC's [5].

This work done incorporate the foundation on which research isto be done to develop low cost, high efficiency solar energyto electricity conversion units for use especially in ruralareas of Sri Lanka where access to the national gridelectricity supply is not available. This would also enable toalleviate poverty and to improve living standards amongstrural communities. .

Figure 1. Absorption spectra of dye solutions extracted from(a) Mangoostien (b) Rabutan (c) Mango (d) Tomato (e) Carrot(f) King coconut (g) Pumpkin (h) Red Banana (i) Beetroot

7

0

0.1

0.2

0.3

0.4

0.5

0 100 200 300 400 500 600 700Voltage (m V)

Curre

nt D

ensity (m

A cm

-2 )

(c)

(e)

(b)

(d)

(a)

(g)

(h)(i)(f)

Figure 2. Current-Voltage characteristics of pigmentsensitized solar cells (a) Mangoostien (b) Rabutan (c) Mango(d) Tomato (e) Carrot (f) King coconut (g) Pumpkin (h) RedBanana (i) Beetroot

Table 1 Characteristics of a few solar cells measuredinitially with different natural dye extracts.

Summary of Results

Ser.No.

Dye Voc(mv)

Isc(mA)

Id(mA/cm2

)

FillFactor(%)

Efficiency(%)

Solvent

01 Mangoostien(a)

556 0.108 0.432 64 0.153 Ethanol

02 Rabutan (b) 504 0.078 0.312 46 0.072 Ethanol03 Mango (c) 567 0.052 0.210 43 0.050 Ethanol04 Tomato (d) 577 0.048 0.194 44 0.052 Ethanol05 Carrot (e) 641 0.038 0.154 35 0.035 Ethanol06 King coconut

(f)256 0.037 0.150 30 0.011 Ethanol

07 Pumpkin (g) 622 0.018 0.076 36 0.017 Ethanol8

08 Red Banana(h)

473 0.008 0.034 52 0.008 Ethanol

09 Beetroot (i) 256 0.004 0.018 37 0.002 Ethanol

Table 2 Summary of Photo Electro Chemical Parameters of DCCC'susing natural dyes from a cross section of plants grown in Sri Lanka(with Decreasing Efficiency Values)

Ser.No.

Dye Voc(mv)

Isc(mA)

Id(mA/cm2

)

FillFactor(%)

Efficiency(%)

Solvent

01 Turmeric 599.1

0.135

0.540 69.03

0.223 Ethanol

02 Mangoostien 565.2

0.111

0.444 65.66

0.165 Ethanol

03 Mangoostien-skin

635 0.087

0.348 69.41

0.153 Ethanol

04 Mangoostien 563.6

0.206

0.412 63.63

0.148 Ethanol

05 Mangoostien-skin

567.5

0.094

0.376 57.38

0.142 Ethanol

06 Mangoostien-skin

593.4

0.081

0.324 68.6 0.131 Ethanol

07 Mangoostien-pulp

631 0.065

0.260 67.73

0.111 Ethanol

08 Venivel 529.9

0.079

0.316 57.59

0.097 Ethanol

09 Orange 627.5

0.05

0.200 73.1 0.091 Ethanol

10 Mangoostien + Orange + Grapes

600.8

0.042

0.168 71.91

0.073 Ethanol

11 Rambutan 504.1

0.156

0.312 45.53

0.072 Ethanol

12 Mangoostien + Orange + Grapes

594.7

0.045

0.180 59.35

0.063 Acetonitrilewith tert

13 Orange 624.9

0.039

0.156 64.6 0.063 Acetonitrile

9

withtert

14 Grapes 498 0.049

0.196 60.28

0.058 Acetonitrilewithtert

15 Rambutan 493.9

0.047

0.188 61.67

0.057 Ethanol

16 Orange 558.3

0.034

0.228 73.86

0.057 Ethanol

17 Rathadun 427.2

0.069

0.276 74.39

0.056 Ethanol

18 Venivalgata 538.4

0.039

0.156 67.37

0.056 Ethanol

19 Orange 619.4

0.039

0.156 55.25

0.053 Acetonitrilewithtert

20 Mango 584.4

0.093

0.186 46.92

0.051 Ethanol

21 Mangoostien-skin

665.5

0.034

0.136 57.03

0.051 Acetonitrilewithtert

22 Grapes 557.8

0.035

0.140 62.86

0.049 Ethanol

23 Spinach 546.5

0.032

0.128 66.43

0.047 Ethanol

24 Bulu 493.7

0.041

0.164 74.44

0.038 Ethanol

25 Mangoostien + Orange + Grapes

608.9

0.026

0.104 56.82

0.036 Acetonitrilewithtert

26 Carrot 641.2

0.077

0.154 35.4 0.035 Ethanol

27 Grapes 538.2

0.024

0.096 61.32

0.032 Acetonitrilewith tert

28 Grapes 566.8

0.015

0.060 72.22

0.025 Ethanol

10

29 Rathadun 415 0.026

0.104 48.58

0.021 Ethanol

30 Beli Flower 490.2

0.019

0.076 49.72

0.018 Ethanol

31 Nelum Seed 497.1

0.022

0.088 39.04

0.017 Ethanol

32 Pumpkin 622.2

0.038

0.076 36.28

0.017 Ethanol

33 KothalaHimbutu

450.1

0.022

0.088 38.87

0.016 Ethanol

34 Tomato 507.5

0.016

0.032 79.69

0.013 Ethanol

35 Aralu 468.9

0.026

0.104 26.83

0.013 Ethanol

36 Purple Makaral 380.7

0.025

0.100 34.13

0.013 Ethanol

37 Walmadata 520.7

0.084

0.336 68.54

0.012 Ethanol

38 Ginger 489.5

0.013

0.052 40.72

0.011 Ethanol

39 King CoconutNut Husk

256.7

0.073

0.146 30.48

0.011 Ethanol

40 Banana 599.2

0.021

0.084 78.77

0.004 Ethanol

41 Red Banana 476 0.016

0.016 48.62

0.004 Ethanol

42 Beetroot 252.7

0.009

0.018 37.54

0.002 Ethanol

43 F.Berry 383.3

0.006

0.024 26.3 0.002 Ethanol

44 B.Onion 331.8

0.002

0.008 32.87

0.001 Ethanol

45 Beli 194.3

0.003

0.012 23.01

0.001 Ethanol

46 PurpleMaakaral

165.1

0.003

0.012 35.56

0.001 Ethanol

47 Adesia Fruit 182.1

0.002

0.008 29.2 0.001 Ethanol

48 Rata Lovi 164.4

0.001

0.004 36.1 0.001 Ethanol

49 Wadukuda Fruit 178.2

0.003

0.012 34.9 0.001 Ethanol

11

50 Goraka 323.3

0.002

0.008 27.91

0.001 Ethanol

3. Research Priorities

The main research concerns for dye- sensitized solar cellsare the reduction of material degradation that leads to poordevice longevity, affordable encapsulation methods to protectagainst environmental degradation , alternatives to dyes assensitizer , and development of solid electrolytes to avoidleakage problems [6].

4. Conclusions

This paper describes an investigation of natural dyes ofplants grown in Sri Lanka as natural photosensitizers.Extracts of Turmeric root and Mangoostein fruit rind haveachieved fill factors, of nearly 70% and 65% respectively, andsolar energy conversion efficiencies of 0.223% and 0.165%respectively, measured with instrumental errors of + or – 1% .

Natural dyes based solar cells appear to be limited by low Vocand Isc. The way forward would be to find different additivessuch as Acetic acid , Hydrochloric acid etc which when dopedin trace amounts would cause alteration of the chemicalstructure of these natural dyes hence to darken their coloursand possibly to result in larger conversion efficiencies.Although natural dyes are still below the presentrequirements, the results are encouraging and may boostadditional studies oriented to the search of new natural dyesensitizers.

Acknowledgements 12

Provision of laboratory facilities to carry out the above -mentioned experiments on DSSC's out the above mentionedexperiments at the Institute of Fundamental Studies (IFS),Hantana, Sri Lanka under the guidance of Professor G.K.R.Senadheera is gratefully acknowledged. Also the support ,advice and guidance for this research work provided byProfessor A.A.P. De Alwis, Dr. B.A.J.K. Premachandra of theUniversity of Moratuwa are also acknowledged with gratitude.

References

1. O Regan, B., Gratzel, M., 1991. A low-cost, high-efficiency solar cell based on dye-sensitized colloidalTiO2 films. Nature 353, 737 – 740.

2. Grätzel, M., 2001. Photo electrochemical cells. Nature414, 338-344.

3. Smested, G.P., 1998. Education and Solar Conversion:Demonstrating electron transfer. Solar Energy Materials & Solar Cells 55, 157-178.

4. Grätzel, M., 2003. Dye-sensitized Solar Cells. Journal ofPhotochemistry and Photobiology C: Photochemistry Reviews(4), 145-153.

5. Hao, S., Wu, J., Huang, Y., Lin, J., 2006. Natural dyes asphoto sensitizers for dye-sensitized solar cells. SolarEnergy 80, 209-214.

6. Hasan,M.S., Roy,S., Corkish,R., 2010. Sustainable Energyin Asia and the Pacific: Emerging Technologies andResearch Priorities in Solar Energy, 28- 33.

13