All the reactives and solvents used where pro analysis quality or better: pyrrole and 4-Bromobenzaldehyde for synthesis were obtained from Sigma-Aldrich; propionic acid, hexane, chloroform, ethanol, methanol, sodium hydroxide, acetone, acetic acid 100%, boric acid, orthophosphoric acid 85%, silica gel 60 230 - 400 mesh ASTM, 0.04 - 0.063 mm porosity, cobalt acetate (II), deuterated chloroform and sodium sulfite were from Merck; potassium chloride was obtained from Riedelde-Haen; tetrabutylammonium perchlorate 99% from Across-Organics; silicone oil from Vetec; dichloromethane from J.T Baker; and ultra pure nitrogen 99.995 from AGA.

Microanalysis was performed on a Fison, Model EA- 1108, elemental analyzer. Proton nuclear magnetic resonance spectra (1H-NMR) were registered in a Brucker Advance 400 MHz spectrometer and processed by Mestrec software. UV-visible electronic spectra were registered in a Scinco S-3100 spectrophotometer, which has a diode bridge detector on 1 cm optical pass quartz cell. The equipment is controlled by the LabPro software through an interface. The spectrum window was 350 to 750 mm. All the measurements were taken in dichloromethane at room temperature.

The monomer films were prepared by thermal sublimation in high vacuum at 1 × 10^–4 Pa of pressure at the same time as the photovoltaic devices. The deposit rate of each monomer was 0.5 Å·s^–1. In order to investigate the optical behavior at the same conditions than the solar cell, the optical properties were determined in thin film solid state (optical density O.D).

The photovoltaic devices were assembled in a multi heterojunction structure as follows: indium tin oxide (ITO)/anode buffer layer MoO₃ and/or CuI/BrPhPZn/ fullerene (C₆₀)//Alq₃/Al. The devices were prepared by sublimation at 1 × 10^–4 Pa. The thin film deposition rates and thickness were estimated in situ through a quartz monitor. Electrical characterizations were performed with an automated I-V tester, in the dark and under sun global AM 1.5 simulated solar illuminations. Performances of photovoltaic cells were measured using a calibrated solar simulator (Oriel 300 W) at 100 mW/cm² light intensity.

The characteristics of the photovoltaic cells were measured using a calibrated solar simulator (Oriel 300 W) at 100 mW/cm² light intensity adjusted with a reference cell (0.5 cm² CIGS solar cell, calibrated at NREL, USA).

The thin film structures are analyzed by X ray diffraction (XRD) by a Siemens D5000 diffractometer using K_α radiation from Cu (λ_Kα = 0.15406 nm).

The complex was synthesized according to Adler Method described in literature [9]. 4-Bromobenzaldehyde (3.9 g, 21 mmol) and pyrrole (1.5 mL, 21 mmol) in 150 mL of propionic acid were heated by reflux. After 30 minutes, the formation of a black precipitate was observed. It was cooled at room temperature, and methanol was added, cooling it by an ice bath at 0˚C during 15 min. Subsequently, it was filtered and the filtrand was washed with distillated water and methanol, dried and later purified by column chromatography (silica/CH₂Cl₂), obtaining a solid and crystalline purple precipitate (34% yield): Elemental analysis: C₄₄H₂₆N₄Br₄ (H₂BrPhP); calculated: C: 56.97, H: 2.83, N: 6.04; found: C: 57.01, H: 2.80, N: 6.02. 1HNMR (400 MHz, CDCl₃) H₂BrPhP: d 8.80 (s, 8H), 8.02 (d, 8H), 7.96 (d, 8H), –2.98 (2H).

H₂BrPhP (2.33 g, 2.5 mmol) and copper acetate monohydrate (1.99 g, 10 mmol) in dichloromethane and ethanol (1:1) were placed in a Schlenk reactor to be purged with argon for 5 minutes and later heated by reflux for 6 hours. The obtained violet solution was cooled at room temperature and its organic phase was extracted by dichloromethane and water. The organic phase was dried with sodium sulfate anhydride which was filtered afterwards. The product that contains the metallated porphyrin was dried before being purified by silica gel column chromatography, and the significant compound was eluted with a mixture of dichloromethane:hexane (3:1). The product obtained is violet colored and has a 65% yield: C₄₄H₂₄N₄Br₄Zn (BrPhPZn).

In order to understand the effect of the different ABL on the OPV cells performances, different characterization techniques have been used for studying the structures ITO/ABL and ITO/ABL/BrPhPZn.

Thin films structures were analyzed by X ray diffracttion (XRD) by a Siemens D5000 diffractometer using K_α radiation from Cu (λ_Kα = 0.15406 nm).

Optical measurements were carried out at room temperature using a Carry spectrometer. The film transmittance was measured at wavelengths of 1.2 to 0.30 μm.

Morphology of the different structures used as anode was observed through scanning electron microscopy (SEM) with a JEOL 7600 F at the “Centre de microcaractérisation de l’IMN, Université de Nantes”. Images in secondary (SEM) and backscattering (BEI) mode have been done. Composition of the films were studied by electron probe micro analysis (EPMA), where the scanning electron microscope was equipped with a BRUKER Quantax X-ray microanalysis system; X-rays were detected by a silicone drift detector.

AFM images on different sites of the film were taken ex-situ at atmospheric pressure and room temperature. All measurements have been performed in tapping mode (Nanowizard III, (JPK Instruments). Classical SiN₄ cantilevers were used (Ultrasharp, AXESSTECH, France).

The average force constant and resonance were approximately 14 N/m and 320 kHz, respectively. The cantilever was excited at its resonance frequency.

An OPV cell needs a transparent conductive electrode which allows the photons to reach the photoactive layer. Here, the glass substrate is coated with 100 nm of a conductive oxide, the indium and tin oxide. The photoactive components are displayed as nm-thin layers. Between the transparent anode and the ED a mono or bi ABL (MoO₃ or MoO₃-CuI) is placed, which increases the photovoltaic device efficiency through an improvement of the band matching at the interface anode/electron donor. The next layer is constituted by the photoactive material or electron donor compound (ED), meso-tetrakis (4-Bromophenyl)zinc porphiryn (BrPhPZn) [10-12]. On this layer a layer of the electron acceptor material, which are generated by the fullerene photoactive material C₆₀, is deposited. Actually, C₆₀ has a high electroafinity value, which makes it a good electron acceptor and also exhibits satisfying conductivity (10^–4·Scm^–1) [13-15]. Afterwards, a tris-8-hydroxyquinoline aluminum, Alq₃, layer is placed, which is a very efficient organometallic complex with great stability and it is used as exciton blocking layer. Finally, the cathode, often is a thin aluminium layer (100 nm), is deposited by thermal evaporation through a mask with 1 mm × 10 mm active areas (Diagram 1).

According to this diagram, sun photons would go through the transparent electrode and would be absorbed into the device by the active organic layer, BrPhPZn, which leads to the generation of the exiton (for the electron hole) [15,16].

The characterization of the cells was carried out by a common method for fotovoltaic cells [17,18], which is based on the determination of a series of parameters such as open circuit voltages (Voc), current density intensity in short-circuit (Isc), the fill factor FF), and the efficiency of the luminous energy conversion. The intersecttion of the I-V curves and the abscissa axis gives the open-circuit voltage value (Voc) and the intersection with the coordinates axis gives the value of the current density in short-circuit (Isc). The maximum power will

Diagram 1. Typical structure of an organic solar cell (ITO/ MoO₃-CuI/BrTPP(Zn) (4-Bromophenyl zinc porphiryn)/ C₆₀/Al.

be the maximum product of the product IxV.

A typical result is shown in Figure 1.

The insertion of a thin anodic buffer layer, ABL, favors the anodic contact between ITO and the ED, which plays an important role in the photovoltaic cells efficiency. One of the most commonly used ABL corresponds to transition metal oxides: MoO₃, V₂O₅, WO₃, that are highly transparent and which have been successfully used to modify the anodic surface. The thickness of such metal oxide is important. A too thin ABL leads to inhomogeneous properties at the interface anode/ED by incomplete surface coating [19], while a too thick layer increases the serial resistance (Rs) and therefore reduces Jsc and FF. We have already shown that, in the experimental conditions used in this work, the optimum thickness of the MoO₃ ABL is 3 nm [20].

It is known that such MoO₃ ABL allows achieving good band matching at the interface anode/ED. Actually, much research has been dedicated to the adjusting of the band matching to reduce the barrier at the interface anode/electron donor through the introduction of an ABL

[20-24].

No much research has been devoted to the effect of the surface energy of the anode on the structure and morphology of the organic films [25,26]. However, the carrier transport properties of the organic semiconductor comprising π-conjugated molecules strongly depend on the molecular order in the films. So it could be of interest to use an ABL which not only improves the band matching at the interface, but also allows some improvement of the order in the absorbing organic layer. Recently, it has been shown that CuI can be used to control the molecular orientation of metal phthalocyanine (MPc) in organic cells based on MPc/C₆₀ [27-29] Therefore, in the present research we extend the study to the effect of CuI on the performances of organic cells based on multi-heterojunction using porphyrin as electron donor through double ABL: MoO₃-CuI.

As it can be seen in the results showed in Table 1, the insertion of the double MoO₃-CuI ABL increases the yield of organic photovoltaic cells (OPV) close to the double of magnitude if compared to the device prepared with a mono layer of MoO₃.

It is important to point out that the improvement by the insertion of a thin CuI layer is strongly related with the improvement of Voc and Jsc while there is no significant evolution of FF.

We show in the Table 1 that, whatever the thickness of the BrPhPZn, the MoO₃-CuI ABL allows improving the solar cells performances. We will discuss more precisely the influence of the BrPhPZn below.

The study of the effect of the thickness of the porphyrin film, BrPhPZn, on the efficiency of the organic photo-

Table 1. Characterization of a photovoltaic cell, ABLs.

voltaic cells (OPV cells) is summed up in Table 2. The device which showed the highest efficiency was the one with a BrPhPZn film thickness of 12.5 nm. For thinner films (10 nm), there is no more improvement of the cells performances. For BrPhPZn layers thick of 50 nm it is observed that the Jsc and Voc values are low, and they increase gradually as the film thickness diminishes. This response could be related to the low conductivity and hole mobility of the compound [19].

The thickness value of CuI_, directly influences the efficiency of the prepared devices, as the results in Table 3 show. It tells that the best performances are obtained with a CuI thickness of 1.5 nm. This thickness allows achieving the best compromise between a small series resistance Rs and an acceptable shunt resistance Rsh. This compromise allows to increase significantly Jsc without much affect the value of Voc.

In order to try to understand the influence of CuI and MoO₃ on the performances of the OPV cells, we have submitted the structures: ITO/CuI/BrPhZn, ITO/MoO₃/ BrPhZn and ITO/MoO₃-CuI/BrPhZn to some specific characterization techniques.

First of all, the BrPhZn layers obtained are amorphous, whatever the anode configurations, with or without ABL.

The absorption of the BrPhZn films (20 nm) deposited onto ITO and the different ABL is reported in Figure 2. Figure 2 shows the expected Soret and Q bands for the metaled complex. The Soret intense absorption band (β) in the 400 nm to 430 nm region, these bands are characteristic for porphyrins and are due to π – π* ligand transitions [30]. Over the 500 nm, 2 bands (Q) can be observed due to zinc insertion in the porphyrin ring, which generates an increase of the symmetry (from D2h to D4h).

We can see that the optical density of films deposited onto CuI is significantly higher than that of films deposited onto ITO. Moreover, the absorption of the film deposited on MoO₃ is less than when it is deposited on ITO, but the introduction of a CuI layer onto MoO₃ increases this absorption.

These results could justify, at least partly, the improvement of the OPV cells performances induced by the double MoO₃/CuI ABL.

In order to determine the effect of the different ABLs on the BrPhZn film conductivity, we have investigated the J-V characteristics of hole-only devices with MoO_3, CuI and MoO₃/CuI ABL. These devices were grown using high work-function electrode buffer layers. The hole only devices were fabricated by replacing the C₆₀ and the BCP EBL with high work function MoO₃, which is well known to be a hole injector (collector) and a blocking electron layer. Hole only devices have been made using the same ITO covered glass substrate than those used to grow OPV cells. After deposition of the ABL, a BrPhZn film thick of 30 nm has been deposited. Then the organic film has been covered with a MoO₃ film thick of 7 nm. Finally aluminium has been used as top electrode. The J-V characteristics of the hole only devices are shown in Figure 3. Charge transport in these devices is limited to holes. Electron injection from the aluminum electrode can be neglected, due to the large electron bar rier due to the MoO₃ interlayer.

The devices using CuI ABL show largest current density at the same driving voltages as compared to the devices using MoO₃ ABL. Clearly the conductivity is enhanced by an order of magnitude when a CuI/MoO₃ ABL is used on the place of the MoO₃ ABL. All the devices display near ohmic transport at low voltage. Actually, in the low voltage range, ohmic contacts can be seen by the linear relation between the current density J and the voltage V at higher current density the slope increases

Tabla 2. Characterization of a photovoltaic cell, thickness of the BrPhPZn film.

Table 3. Characterization of a photovoltaic cell, thickness of the CuI film.

which corresponds to a transition towards Space Charge Limited Current (SCLC) which is often obtained in such devices [30].

The studies of the layer morphology by scanning electron microscopy (SEM) and atomic force microscopy (AFM) allow to determine the rugosity and superficial characteristics of the donor organic layer, BrPhPZn, which will depend on the ABL layer in contact with the latter.

The visualization of the layers under the scanning microscope allows to see butterfly-shaped “features”, which are scattered over the surface See Figure 4.

In Figure 5, on the other hand, we can see that the

concentration or population density of these “butterfly-like” morphologies is higher with CuI, which can be proved with the AFM. In Figures 6 and 7 we have measured the rugosity, which can be clearly seen in tridimensional images with the AFM, the Rms over MoO₃/CuI = 15 nm, and over MoO₃ el Rms = 16 nm, that is to say that is more or less the same rugosity, what changes is the density of these “butterfly” kind of formations. The cross sections make it possible to see the different densities of rugosity.

According to these results, it can be seen that the morphological determination by SEM and AFM allows for a clear sight of the influence of the blocking layer (MoO₃ or CuI) over the growth of the organic layer. This differ-

ent growth allows the understanding of the differences in the conductivities of the layers deposited over CuI. The increase in the yield of the cells that used CuIO as ABL would be because of the increase in the conductivity of the donor layer for the introduction of CuI.

According to these results, it can be seen that the morphological determination by SEM and AFM allows for a clear sight of the influence of the blocking layer (MoO₃ or CuI) over the growth of the organic layer. This different growth allows the understanding of the differences in the conductivities of the layers deposited over CuI. The increase in the yield of the cells that used CuIO as ABL would be because of the increase in the conductivity of the donor layer for the introduction of CuI.

Another limiting factor affecting power conversion efficiency is the small diffusion length of the exciton before recombination. Typical diffusion lengths of exciton are in the range of 10 nm. So, it is necessary to improve the interface area between the electron donor and the electron acceptor in order to improve the probability of charge separation. A possible approach to achieve a large interface area is the use of a quite rough active organic layer. In the present case, the higher density of rough feature in the case of CuI ABL can participate to the improvement of the photocurrent.