Essentially the most stunning results of conductivity enhancement of the PEDOT:PSS has been achieved by a number of spin coatings of the PEDOT:PSS and added Cu NP dispersion. The outcomes are summarized in Desk 1.
The doubling of the PEDOT:PSS layer alone decreases the sheet resistance by greater than tenfold, from 1MΩ/sq (pattern 1) to 90kΩ/sq (pattern 3) and 85kΩ/sq (pattern 4) relying on the spinning velocity. A deposition of a 3rd PEDOT:PSS layer diminished the sheet resistance additional to 60kΩ/sq (pattern 6) and even to 28kΩ/sq (pattern 7) relying on the spinning velocity. Much more stunning is the discovering that the thickness of the PEDOT:PSS coated twice (twice all the coating cycle together with the annealing step) and thrice doesn’t change the ultimate thickness appreciably. The thickness of a triple deposition at 1500 rpm, 3000 rpm and 1500 rpm offers a remaining thickness of 85 nm. If the depositions could be strictly additive it might be 65 + 29 + 65 = 159 nm. In fact the spinning velocity impacts the ultimate thickness. Thus, for instance a triple PEDOT:PSS deposition with consecutively greater speeds: 1500 rpm, 2000 rpm, and 3000 rpm offers a complete movie thickness of 72 nm, which is barely 7 nm thicker than a single layer PEDOT:PSS deposited at 1500 rpm and much lower than the additive variety of 65 + 56 + 29 = 150 nm. As talked about within the introduction throughout the spinning a vertical dephasing of the PEDOT and PSS elements takes place. The moist movie comprises the conductive PEDOT strips on the backside of the movie and a PSS-rich answer on the higher half. Throughout the subsequent smooth bake at 120 °C these segregated layers have inadequate time for remixing. The reason for the phenomenon is given in Fig. 4, the place the PSS-rich prime layer and the PEDOT-rich backside layer are proven in a schematically simplified manner. There’s, in fact, no sharp transition between the 2 phases, however a gradual one, with excessive focus of PSS on the prime and lowest on the backside with the reverse conduct of the PEDOT. As identified within the introduction, throughout the spin deposition of PEDOT:PSS the hydrophobic PEDOT-rich part settles on the backside of the movie and the hydrophilic PSS-rich part on the prime of the PEDOT:PSS movie. Throughout the second spin coating the hydrophilic PSS prime portion of the movie will get uncovered to water and is partially supplanted by PEDOT-rich part.
Rationalization of how a number of PEDOT:PSS coatings improve {the electrical} conductivity of the movie. As defined within the textual content, the thicknesses of a double layer, and a triple layer enhance solely very barely over the thickness of a single layer. The a number of coatings assist to supplant the PSS prime portion of the layer with PEDOT:PSS ribbons.
The identical occurs throughout the third spin-coating (samples 6–10) with additional reducing of the sheet resistance to 12kΩ/sq (pattern 8). Consequently, the PEDOT-rich part grows on the expense of the PSS-rich part. Because the conductive PEDOT-rich part will increase after repeated coating cycles, {the electrical} conductivity of the PEDOT:PSS movie is considerably improved. Throughout the second and third spin coating the moist movie helps take away the PSS materials and supplant it with conductive PEDOT part. Consequently, the conductivity will increase considerably. The spinning velocity has an attention-grabbing impact on the conductivity of the PEDOT:PSS movies. Pattern 8 differs from pattern 6 solely within the spinning velocity of the threerd PEDOT:PSS layer of 3000 rpm as an alternative of 1500 rpm. The upper velocity throughout the third PEDOT:PSS deposition lowers the sheet resistance by an element of 5 from 60kΩ/sq to 12kΩ/sq. The reason for this impact is in keeping with rationalization given in Fig. 4. The upper velocity of the third deposition doesn’t lower the thickness of the PEDOT:PSS movie as a result of the thickness of the general movie is given mainly by the primary coating whereas the following depositions serve solely to densify the PEDOT part on the backside portion of the layer forcing this elevated connectivity between the PEDOT ribbons.
To confirm this assumption, the sheet resistance of a single PEDOT:PSS layer spun at 1500 rpm has been discovered to be 230kΩ/sq, (pattern 2) i.e. 77% decrease than the one with PEDOT:PSS spun at 2000 rpm (pattern 1). Two layers of PEDOT:PSS each at spun at 3000 rpm (pattern 4) resulted in 85kΩ/sq roughly in the identical vary (80kΩ/sq) as 2 layers spun at 1500 rpm (pattern 5). Three layers of PEDOT:PSS (pattern 7), with all three coatings spun at 3000 rpm, lead to Rsq = 28kΩ/sq. This demonstrates a tradeoff between the extra denser part of PEDOT attained at greater velocity charges and the thickness of the layer. The denser PEDOT part decreases Rsq whereas a thinner PEDOT:PSS layer tends to extend Rsq. Therefore, the optimum answer consists in making the most of each elements by spinning the primary two PEDOT:PSS layers at a decrease price to safeguard a thicker polymer movie and add a 3rd layer spun at greater velocity to succeed in greater densification of the PEDOT part. Samples 8, 9, and 10 present additional mixtures of the spinning velocity mixtures. Pattern 9 with a sequence of spinning speeds (1500 rpm, 2000 rpm, 3000 rpm) reveals a sheet resistance of 18kΩ/sq. Pattern 10 with the spinning velocity mixture (1500 rpm, 3000 rpm, 1500 rpm) yields a sheet resistance of 21kΩ/sq. Pattern 8, 9, and 10 present that it’s advantageous to carry out the primary two depositions at low velocity to supply a thicker PEDOT:PSS layer and use excessive velocity throughout the third deposition to realize excessive PEDOT densification on the backside of the layer.
To see the affect of a number of coatings, we’ve got elevated the variety of sequential PEDOT:PSS coatings to 6 coatings, such that the pattern with n coatings shares precisely the identical (n-1) PEDOT:PSS layers because the pattern with (n-1) coatings. On this explicit sequence, the 1 × layer PEDOT:PSS pattern is pattern 2 deposited at 1500 rpm, the second pattern is pattern 3 with two layers deposited at 1500 rpm and 2000 rpm, respectively. Every extra PEDOT:PSS coating has been deposited at 3000 rpm to maximise the conductivity. The samples with 4 × PEDOT:PSS, 5 × PEDOT:PSS, and 6 × PEDOT:PSS are listed in Desk 1 as pattern 22, 23, 24, respectively.
The outcomes of the sheet resistance for the a number of coatings are given in Desk 2 and plotted in Fig. 5.
The sheet resistance, Rsq, of the a number of PEDOT:PSS coatings as a features of the variety of coatings, n. An exponential dependence between Rsq(n) and n is discovered and could be expressed as Rsq(n) = 230 exp[− (n − 1) 1.28] kΩ.
From Fig. 5 an exponential dependence of the sheet resistance as operate of the variety of PEDOT:PSS coatings has been discovered. The exponential trendline in Fig. 5 could be fitted by the next expression Rsq(n) = 230 kΩ × exp[− (n − 1) × 1.28]. Thus, it may be seen that additional PEDOT:PSS coatings would offer diminishing returns. We are going to revisit the a number of coatings of PEDOT:PSS within the context of the effectivity of doping by Cu NP of a number of PEDOT:PSS coatings in decreasing the general sheet resistance.
In our strategy so as to add Cu NP to the movie we’ve got chosen so as to add Cu NP on the prime after the soft-bake of the spin-coated PEDOT:PSS. In distinction to Ag and Au NPs30,32, which could be added to the PEDOT:PSS dispersion and spin coated along with it, the dispersion of Cu NP is tough and ineffective46 as a result of Cu is well oxidized at ambient temperatures. In reality, our experiments confirmed this evaluation. After mixing 0.5 mg/ml Cu(60 nm) NP with PEDOT:PSS in the identical aqueous answer to deposit a single PEDOT:PSS layer doped with Cu NP, the sheet resistance of 1 PEDOT:PSS layer doped in such a manner with Cu NP, elevated barely from 1MΩ/sq to 1.2 MΩ/sq. (This pattern just isn’t being listed within the Desk 1, as a result of such step will increase the sheet resistance as an alternative of reducing it). When integrated Cu NP in an aqueous answer along with PEDOT:PSS, the oxidized Cu NPs trigger much more isolation between the PEDOT ribbons. Nevertheless, when spinning the PEDOT:PSS dispersion with the Cu NPs 3 times, the sheet resistance of the triple PEDOT:PSS(Cu NP) dropped to 44.4KΩ/sq. (This pattern can be not listed within the Desk 1.). This end result signifies that among the Cu NP managed to keep away from oxidation and enhanced the conductivity by serving to bridge electrically the PEDOT ribbons. In fact, such issues would have been averted, when as an alternative of Cu NP, Au or Ag NPs would have been used. Nonetheless, since Cu NPs are extra economical and extra fascinating due to their confirmed ReRAM switching properties, experiments have been continued with Cu NP being now dispersed in ethylene and deposited on prime (to keep away from any contact with moisture) of an annealed PEDOT:PSS.
Desk 1 reveals that the addition of dispersion of Cu NP on the prime of the PEDOT:PSS additional dramatically will increase the conductivity of the movie. A single layer of PEDOT:PSS with 0.2 mg/ml Cu(25 nm) (sample11) lowers the sheet resistance from 1MΩ/sq to 1.2kΩ/sq. The mixture of two layers of PEDOT:PSS with Cu deposited final (pattern 12) offers a sheet resistance of 467 and 390 Ω/sq for 25 nm sized and 60 nm-sized Cu NP (pattern 15), respectively, indicating {that a} bigger measurement of Cu NPs is more practical in reducing the sheet resistance. A mix of three layers of PEDOT:PSS and with a prime layer of Cu NP (60 nm) (pattern 16) yields 173 Ω/sq. As could be seen from pattern 17, the extra two Cu NP coatings resulted in a marginal enchancment from 173 to 162 Ω/sq. Pattern 18 is much like the pattern 17, nevertheless in pattern 17 the Cu dispersion on prime of the three layers of PEDOT:PSS has been coated 3 times with Cu NP. In samples 14 and 18 we investigated the impact of Cu NP coating after deposition of each PEDOT:PSS layer. The ensuing sheet resistance is reasonably excessive of 1.2kΩ/sq and 1.0kΩ/sq, respectively. The seemingly purpose for the excessive resistance is that the efficacy of the Cu NP doping as a result of Cu NP deposition on the primary two PEDOT:PSS layers is degraded by the oxidation when aqueous PEDOT:PSS is being deposited on prime of the Cu NP coating. It might probably, thus, be seen that the results of a number of spin coatings of PEDOT:PSS and of Cu NP are not at all additive and rely sensitively on the sequence of coatings. In some instances they work synergistically, however in different mixtures they are often detrimental to the reducing of sheet resistance.
The a number of coatings of PEDOT:PSS, nevertheless, seem like helpful in all circumstances envisioned right here, since they enhance the amount of the PEDOT-rich backside layer and supply a bigger potential for conductivity enhancement by steel NP doping, together with the Cu NPs. On the identical time, as a result of nature of particle diffusion, the best focus of Cu NPs stays on the prime portion of the PEDOT:PSS movie that’s initially probably the most resistive a part of the movie and rendering it considerably extra conductive as proven in Fig. 6.
Optical microscope image of PEDOT:PSS with and with out Cu NP. (a) single coating of PEDOT:PSS deposited at 1500 rpm with no Cu NP. (b) a triple coating of PEDOT:PSS doped with 2 mg/ml of Cu NP. (c) a single coating of PEDOT:PSS deposited at 1500 rpm with Cu NP. (d) a triple coating of PEDOT:PSS deposited at 1500 rpm with 0.5 mg/ml of Cu NP.
In Desk 3 the interplay between the variety of PEDOT:PSS coatings and doping with Cu NP and its affect on Rsq is elucidated.
It may be seen that the relative discount attributable to addition of Cu NP is strongest for 3 PEDOT:PSS layer, the place the discount issue is greater than two orders of magnitude. In case of 6 PEDOT:PSS layers the discount issue is just one order of magnitude, whereas the discount issue for a single PEDOT:PSS layer is lower than 4. This explicit conduct will develop into clear when the morphology of the doped and undoped PEDOT:PSS movies with a number of coatings will likely be mentioned additional beneath.
The presence of Cu NP on the prime of the PEDOT:PSS has the added benefit of creating Cu+ obtainable for the resistive switching conduct as noticed in lots of ReRAMs together with in Cu/TaOx/Pt47 and Cu/P3HT(GNP)/Au48 reminiscence cells. As seen from Desk 1 bigger measurement of Cu NP, 60 nm vs 25 nm, helps additionally to boost the electrical conductivity of the movie. Samples 19, 20, and 21 (samples with three PEDOT:PSS layers with Cu (60 nm) NP deposition) and samples 8, 9, and 10, additionally with three PEDOT:PSS layers with no Cu deposition present comparable tendencies with respect to the spinning speeds of the PEDOT:PSS depositions, confirming that decrease spinning velocity of the primary two depositions adopted by a excessive velocity deposition of the third layer yield the bottom sheet resistance.
The synergistic impact of a number of coatings, Cu dispersion and the concomitant soft-bakes could be additional optimized to realize even greater conductivities of the movie. From Desk 1 it may be additionally seen that the focus of Cu NP dispersion has a slight enhancement on the enhancement of the conductivity. The comparability of 0.2 mg/ml (pattern 12) and 0.5 mg/ml (pattern 13) reveals that greater focus of Cu NP within the dispersion results in a barely decrease sheet resistance.
Because the methodology of Cu NP doping as a floor layer together with the a number of PEDOT:PSS coatings proved to be very efficient, it was tried to additional optimize the Cu deposition course of. In Desk 2 we present the outcomes of samples with a triple PEDOT:PSS layer and totally different Cu NP deposition recipes with subsequent smooth bakes at numerous temperatures are proven. For samples 26–29 the spinning velocity of the Cu NP dispersion answer has been lowered from 1500 to 500 rpm within the try to extend the thickness of the Cu dispersion coating and thus the Cu focus on prime of the threerd PEDOT:PSS layer. Then the quadruple layer methods have been annealed between room temperature and 120 °C.
From Desk 4 it may be seen that the change of the sheet resistance as operate of sentimental bake at temperatures from 60 °C to 120 °C is insignificant. The smooth bake at 60 °C appeared to enhance the sheet resistance solely barely, however the subsequent smooth bakes at 90 °C and 120 °C degraded the sheet resistance by 6% and 21%, respectively.
Evaluating samples 4 and 5 with two PEDOT:PSS layers it may be seen that the utmost spinning velocity had a slight impact on the conductivity: the conductivity for the utmost velocity of 3000 rpm for each PEDOT layers is 85 kΩ/sq and rotating each layers at 1500 rpm resulted in 80 kΩ/sq, whereas rotating the primary PEDOT:PSS layer at 1500 rpm and the second at 2000 rpm resulted in a sheet resistance of 90 kΩ/sq (pattern 3). Two PEDOT:PSS layers at 2000 rpm (not listed within the Desk 1) has a sheet resistance of 99.7 kΩ/sq. This end result signifies that the sheet resistance will increase with the spinning velocity brought on by the thinning of PEDOT:PSS layer with greater spinning speeds.
The 2 most promising approaches, i.e. the double PEDOT layer and doping with Cu NP have been mixed in two separate configurations. Within the first configuration, the Cu NP have been spun after the deposition of the primary PEDOT layer at 2000 rpm. and the second configuration consisted in a double layer PEDOT:PSS deposition, adopted by one other deposition of the Cu NP. The primary configuration yielded a sheet resistance of 1.2KΩ/sq (pattern 11) and the second resulted within the very low sheet resistance of solely 467 Ω/sq (pattern 12).
Morphology of the PEDOT:PSS movies
To characterize the PEDOT:PSS movies morphologically optical microscope photos of the PEDOT:PSS movies for a number of coatings have been taken with and with out Cu NP in addition to Atomic Pressure Microscopy (AFM) to evaluate the floor roughness of the movies. In Fig. 6 the optical microscope photos of 1-coating, and 3-coatings of PEDOT:PSS movies with and with out Cu NP are proven. It may be seen that a number of coatings enhance the density of PEDOT:PSS particles seen within the photos as small white dots. The triple coating of PEDOT:PSS has a extra pronounced whitish tint in contrast with the only PEDOT:PSS coating, indicating the next density of the PEDOT:PSS composites. The one PEDOT:PSS layer shows wider gaps between the PEDOT:PSS particles rendering the optical picture distinction darker. In case of PEDOT:PSS movies doped with Cu NP one can distinguish solely a part of them recognizable by an orange-reddish glow. They are often simply noticed due to the their comparatively giant measurement between 60 and 80 nm. If one tilts the pattern solely so slightly below the microscope, different Cu NP could be seen, when the orientation of their sides trigger seen reflection of the sunshine into the digital camera. The Cu NP seem to settle in locations the place the density of PEDOT:PSS particles is low. Thus, they have a tendency to fall within the cavities between the PEDOT:PSS particles.
Further data could be gained from AFM photos of the surfaces of the corresponding PEDOT:PSS movies. In fact, the AFM photos are insensitive to the fabric variations. In Fig. 7, AFM photos of chosen PEDOT:PSS movies are proven. In Desk 5, the foundation imply sq. (rms) floor roughness of the AFM plots for the respective samples, has been summarized.
AFM photos of PEDOT:PSS of a single, triple, and sixfold coatings of PEDOT:PSS with and with out Cu NP. The scanned space of all of the plots is 500 nm × 500 nm. The foundation imply sq. floor roughness, srrms, can be indicated for the respective floor floor. (a) a single PEDOT:PSS no Cu NP; srrms = 2.01 nm, (b) a triple PEDOT:PSS no Cu NP; srrms = 2.42 nm, (c) a sixfold PEDOT:PSS no Cu NP; srrms = 2.37 nm, (d) a single PEDOT:PSS with Cu NP; srrms = 4.10 nm (e) a triple PEDOT:PSS with Cu NP; srrms = 2.29 nm, (f) a sixfold PEDOT:PSS with Cu NP; srrms = 2.90 nm.
It may be seen that for one coating of PEDOT:PSS the floor roughness with Cu NP is far greater than with out Cu NP. That is believable because the movie thickness of 1xPEDOT is ca. 65 nm and the scale of Cu NP is between 60 and 80 nm. Thus, inevitably, some Cu NP are sure to stay out, contributing to the next floor roughness. Nonetheless, the relative low density of PEDOT:PSS particles are capable of accommodate the Cu NP is the empty areas between them as solely a small higher a part of the Cu NP is protruding. In case of 3xPEDOT:PSS the movie thickness is 85–90 nm and the density of PEDOT:PSS nonetheless low sufficient permitting enough free house to accommodate the Cu NP solely within the movie. From Desk 5 it may be seen that the floor roughness with Cu NP is now decrease than with out them (in distinction to the case with one coating and 6 coatings). The decrease floor roughness with Cu NP is more likely to be brought on by the truth that now the PEDOT:PSS movie thickness is bigger than the most important Cu NP and the density of PEDOT:PSS composite particles is low sufficient to accommodate Cu NP solely inside the movie. In case of 6xPEDOT, the movie thickness is greater than for 3xPEDOT:PSS however the density of PEDOT:PSS is now so excessive that there are fewer empty areas to accommodate the Cu NP, thus the floor roughness with Cu NP is considerably greater than with out them (see Desk 5). The leads to Desk 5 correlate completely with the findings of the sheet resistance’s dependence on the variety of PEDOT:PSS coatings and the discount issue of the sheet resistance as a result of addition of Cu NP, proven in Desk 3. The Cu NP have the strongest, relative, affect when the movie thickness is equal or bigger than the scale of Cu NP and the movie can provide sufficient empty house between the PEDOT:PSS particles to accommodate the Cu NP solely between them.
Along with the doping methods demonstrated right here, different promising enhancement strategies embrace doping of PEDOT:PSS with Au or Ag NPs or remedy of PEDOT:PSS with acids comparable to HNO3, H2SO4. The optimum mixture will rely on the combination problems with PEDOT:PSS electrodes with doable PEDOT:PSS switching layers right into a remaining natural ReRAM cell. Though the manufacturing of excessive conductivity of the movies is a crucial goal, the switching efficiency of a purely natural ReRAM reminiscence cell is much more essential, rendering the method integration and optimum mixture of enhancement methods, a non-trivial job.
