Hyo Seok
Kim
,
Han Seul
Kim
,
Seong Sik
Kim
and
Yong-Hoon
Kim
*
Graduate School of Energy, Environment, Water, and Sustainability, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, Korea. E-mail: y.h.kim@kaist.ac.kr; Fax: +82-42-350-1710; Tel: +82-42-350-1717
First published on 10th October 2014
It was recently shown that nitrogen-doped graphene (NG) can exhibit both p- and n-type characters depending on the C–N bonding nature, which represents a significant bottleneck for the development of graphene-based electronics. Based on first-principles calculations, we herein scrutinize the correlations between the atomic and electronic structures of NG and particularly explore the feasibility of converting p-type NG with pyridinic, pyrrolic, and nitrilic N atoms into n- or bipolar type by introducing an additional dopant atom. Of the nine candidates B, C, O, F, Al, Si, P, S, and Cl, we find that B-, Al-, and P-codoping can anneal even relatively large vacancy defects in p-type NG. It will be also shown that, while the NG with pyridinic N can be converted into the n-type via codoping, only a bipolar type conversion can be achieved for the NG with nitrilic or pyrrolic N. The amount of work function reduction was up to 0.64 eV for the pyridinic N next to a monovacancy. The atomistic origin of such diverse type changes is analyzed based on Mulliken and crystal orbital Hamiltonian populations, which provide us with a framework to connect the local bonding chemistry with the macroscopic electronic structure in doped and/or defective graphene. Moreover, we demonstrate that the proposed codoping scheme can recover the excellent charge transport properties of pristine graphene. Both the electronic type conversion and conductance recovery in codoped NG should have significant implications for the electronic and energy device applications.
Carrying out first-principles density functional theory (DFT) and DFT-based non-equilibrium Green's function (NEGF) calculations, we herein systematically investigate the atomistic origins of p-type character in NG with pyridinic, pyrrolic, and nitrilic N and the feasibility of achieving robust n-type graphene by incorporating a codopant atom. Graphene codoped with N and B or N and P has already been synthesized and shown to improve the ORR performance,28,29 but the atomistic details and mechanism of the synergistic effects associated with B, N- and P, N-codoping have not yet been understood. We first consider the energetic feasibility of introducing an additional atom into various p-type NG defect sites and show that B, Al, Si and P atoms can structurally anneal even relatively large vacancy defects next to the pyridinic, nitrilic, and pyrrolic N atoms. We find that, except for Si, the B, Al, and P codoping can convert the pyridinic NG into n-type and the nitrilic and pyrrolic NG into bipolar type, and thus effectively eliminate the p-type character of NG. Based on the Mulliken and crystal overlap Hamilton population (COHP) analyses, we further establish the basis to understand how the macroscopic electronic type change in NG is induced from the atomistic bonding viewpoint. Finally, we will demonstrate that an additional benefit of the codoping approach is the recovery of an excellent charge transport capacity of pristine graphene, in line with the recovery of the sp2 bonding network upon the healing of vacancy defects.
Φ = Vvac − EF, | (1) |
(2) |
Transmission functions T(E) were calculated using the DFT-based NEGF method,35 as implemented in TranSIESTA.36 We used the periodic cell composed of six dimer lines in the transport normal direction (14.80 Å) and sampled 66 points. In the transport direction, we used eight and two zigzag chains to model the channel and electrode regions, respectively, and the surface Green's functions were obtained for the corresponding electrode models sampled with the 25 points (see ESI Fig. S1†). For calculating T(E), the energy was scanned from −1.0 eV to 1.0 eV with respect to EF with the 0.001 eV resolution. In many cases, we have crosschecked the validity of our results using SeqQuest and our in-house NEGF code.37,38
In Table 1, we show the zero-temperature formation energies of the V1-Npy, V2-Npy, Nnit, and Npyrr structures calculated according to
Ef(NG) = (EN-doped + xμC) − (Epristine + yμN + zμH), | (3) |
Number of N atoms | Formation energy [eV] | |||
---|---|---|---|---|
V1-Npy | V2-Npy | Nnit | Npyrr | |
1 | 5.14 | 7.44 | 7.66 | — |
2 | 4.48 | 5.53 | — | — |
3 | 2.76 | 4.77 | — | 9.06 |
4 | — | 2.90 | — | — |
An important objective of the present work is to provide an atomistic understanding of the nitrogen bonding nature in NG and its modification upon codoping. For this purpose, we will demonstrate that the combination of band structure, DOS, Mulliken population, and COHP43 provides systematic and detailed information that guides the route toward the desired n-type or bipolar conversion. The COHP is defined by the DOS multiplied by the Hamiltonian of the corresponding element, and the negative COHP values (–COHP) give positive and negative signs for the bonding and antibonding states, respectively. The DOS, COHP normalized by the number of bonds and the band structure data of Ngr, V1-Npy3, V2-Npy4, Nnit, and Npyrr are shown in Fig. 1, and their electronic structures are summarized in Table 2.
Structure | l C–N [Å] | Φ [eV] | Doping type | n [1012 cm−2] |
---|---|---|---|---|
Pristine | 1.43 | 4.23 | — | — |
Ngr | 1.42 | 3.86 | n | 11.31 |
V1-Npy3 | 1.35 | 4.48 | p | −9.31 |
B + V1-Npy3 | 1.41 | 3.84 | n | 16.03 |
Al + V1-Npy3 | 1.38 | 3.86 | n | 15.57 |
Si + V1-Npy3 | 1.38 | 3.84 | n | 12.82 |
P + V1-Npy3 | 1.40 | 3.93 | n | 10.58 |
V2-Npy4 | 1.35 | 4.32 | — | — |
B + V2-Npy4 | 1.42/1.39 | 3.88 | n | 10.52 |
Al + V2-Npy4 | 1.40 | 3.80 | n | 9.80 |
Si + V2-Npy4 | 1.40 | 3.86 | n | 11.48 |
P + V2-Npy4 | 1.39 | 3.86 | n | 5.56 |
Nnit | 1.35 | 4.53 | p | −11.68 |
B + Nnit | 1.46 | 4.13 | — | — |
Al + Nnit | 1.49 | 4.13 | — | — |
Si + Nnit | 1.47 | 4.34 | p | −8.82 |
P + Nnit | 1.47 | 4.11 | — | — |
Npyrr | 1.41 | 4.61 | p | −19.36 |
B + Npyrr | 1.51 | 4.20 | — | — |
Al + Npyrr | 1.45 | 4.15 | — | — |
Si + Npyrr | 1.47 | 4.22 | — | — |
P + Npyrr | 1.48 | 4.21 | — | — |
The doping type of graphitic NG is n-type, i.e., ED is located below EF (Fig. 1a, middle and bottom panels). This results from the electron transfer from the Ngr atoms to the π* states of graphene. Visualizing the Mulliken charge populations as in Fig. 2a, we can observe that the electrons donated from Ngr are distributed throughout the graphene basal plane or become mobile charges. The amount of donated charges extracted from the population analysis is 0.402 electron per nitrogen atom, which is in excellent agreement with the experimental estimation.44 It results in the upward shift of EF from ED of the pristine graphene case to the graphene conduction band or the lowering of the work function by 0.37 eV. The calculated charge-carrier density of 11.31 × 1012 cm−2 for our model with an N atom doping ratio of 0.62% N atoms per C atom (N atom density of 2.31 × 1013 cm−2) is in good agreement with the experimental estimate of 5.42 × 1012 electrons per cm2 for the 0.34% N atoms per C atom doping. The COHP curve shows that these impurity resonant states have antibonding characters, being identified as the strong negative COHP peaks right above EF (thus ED). The energetic locations of the antibonding COHP peaks are EF + 0.07 eV and EF + 0.45 eV. These peak locations depend on the supercell size, or the doping ratio, and are expected to converge to an experimentally observed single peak at about EF + 0.14 eV.27,45
On the other hand, the electronic structures of the other three V1-Npy3, Nnit, and Npyrr NG are p-type or EF has been shifted downward into the valence bands (Fig. 1b, d and e, middle and bottom panels). This should mainly result from the presence of vacancy defect states (whose LDOS are shown in Fig. 1b, d and e, top panels), which behave as acceptor states with missing π-electrons.24,25,27 Note that in these cases the impurity states appear in the COHP plots as strong antibonding peaks right below EF (thus ED). Interestingly, the V2-Npy4 case shows a bipolar character (ED nearly coincides with EF).27 To understand such discrepancies, we analyzed the Mulliken populations as shown in Fig. 2b–d for V1-Npy3, V2-Npy4 and Npyrr, respectively (the Nnit diagram is similar to Npyrr and is not shown). They show that the spatial range of charge redistribution around the nitrogen-vacancy complexes discriminates the V1-Npy3 and Npyrr cases from the V2-Npy4 counterpart: we observed the depletion of electrons throughout the entire graphene basal plane in the former, but rather localized charge depletion around the nitrogen-vacancy site in the latter (compare Fig. 2c with Fig. 2b and d). Related with these charge transfer characters, the impurity states strongly hybridize with the carbon 2pz orbitals or graphene π states (existence of an antibonding C 2pz–N COHP peak) in V1-Npy3 (Fig. 1b, COHP panel, shaded region near EF), whereas the hybridization is very weak (absence of an antibonding C 2pz–N COHP peak) in V2-Npy4 (Fig. 1c, COHP panel, shaded region at E − EF ≈ 0.5 eV).
We now move on to consider the possibility of codoping the p-type NG. Since the presence of a vacancy defect is the precondition for the existence of V1-Npy3, V2-Npy4, Nnit, and Npyrr sites, we focused on examining whether the vacancy defects can be healed by introducing a codopant. In addition to C, eight light elements, B, O, F, Al, Si, P, S, and Cl, were considered as codoping candidates. For the fully optimized structures X + V1-Npy1–3, X + V2-Npy1–4, X + Nnit, and X + Npyrr (X = B, C, O, F, Al, Si, P, S, and Cl), we have calculated the formation energies,
Ef(X + NG) = (Ecodoped + xμC) − (Epristine + yμN + zμH + μX), | (4) |
Erf(X + NG) = Ef (X + NG) − Ef (NG), | (5) |
Fig. 3 Relative formation energies of introducing an additional dopant atom into V1-Npy3 (black squares), V2-Npy4 (red circles), Nnit (blue diamonds), and Npyrr (green triangles). |
First, we find that, except for O + Npyrr, both Nnit and Npyrr defects can be annealed by all of the above codoping elements (negative Erf), because the formation energies of Nnit and Npyrr are rather large to begin with (Table 1) and their relatively large vacancies can easily accommodate an additional dopant atom. However, incorporating an additional dopant atom into the energetically more favorable pyridinic NG was found to be rather difficult: it was determined that the V1-Npy3 and V2-Npy4 defects can be annealed with only a B, F, Al, Si, or P atom. We found that the energetically less favorable pyridinic NG with lower N concentrations can also accommodate these elements (see ESI Fig. S2 and S3†). The results indicate that an Npy to Ngr conversion scheme solely based on the carbon source24 will be limited in that di- or larger vacancies cannot be annealed. Although F codoping of p-type NG might be energetically feasible, it is found that the F atom cannot structurally passivate the vacancy defects in the V1-Npy3 and V2-Npy4 cases and accordingly the electronic structures of F + V1-Npy3 and F + V2-Npy4 still maintain the p-type and bipolar characters, respectively (see ESI Fig. S4†). This leaves B, Al, Si, and P as the codoping candidates for the n-type conversion of p-type NG.
We first focus on the pyridinic NG codoped with B, Al, Si, and P, whose fully relaxed geometries and electronic structures are shown in Fig. 4. The C–N bond lengths of pyridinic NG are elongated when the vacancy is annealed by the codopant (Table 2). We further found that the trend of energetic stability of codoped NG (Fig. 3) is strongly related with the planarity of their geometries. For example, B can anneal the V1-Npy3 defect without protrusion (Fig. 4a), which makes its Erf larger than those of Al, Si, and P. Similarly, Al can anneal the V2-Npy4 defect in a fourfold-coordinated planar configuration (Fig. 4f), which makes it an energetically more favorable codopant than B, Si, and P.
In terms of electronic structures (Fig. 4, bottom panels, and Table 2), we find that all the X + V1-Npy3 and X + V2-Npy4 cases (X = B/Al/Si/P) become n-type, or ED is now located below EF. The passivation of V1-Npy3 and V2-Npy4 vacancy defects by codoping of B/Al/Si/P allows the charge transfer from the N atoms to the graphene π* states (Fig. 2e and f) as in the case of Ngr (Fig. 2a). The alteration of the electronic type change is also easily identified in the COHP data, in which we observe strong antibonding C–N peaks that are pinned near EF or the downward shift of ED below EF. The reduction of work function for V1-Npy3 and V2-Npy4 was up to 0.62 eV and 0.52 eV upon codoping of B/Si and Al, respectively.
To further understand the microscopic mechanisms of electronic type conversion in X + V1-Npy3 and X + V2-Npy4, we have decomposed the C–N COHPs into different orbital contributions (Fig. S5†) and found that the C–N antibonding COHP peaks mostly come from the C 2pz–N antibonds. Namely, the insertion of a B/Al/Si/P atom into V1-Npy3 or V2-Npy4 and the subsequent annealing of vacancy allows the N atoms to directly couple with the sp2 C network and behave like graphitic N. Whereas the amount of downward shift of ED tends to decrease with the lowering of N concentrations, we have confirmed that the B, Al, Si, and P codoping of V1-Npy1–2 and V2-Npy1–3 in general changes their electronic type to n- or at least bipolar ones (see ESI Fig. S6–S10†).
Compared with the pyridinic cases, the nitrilic and pyrrolic NG show a different type of conversion behavior and thus become interesting comparative systems (Fig. 5 and Table 2). The C–N bond lengths in nitrilic and pyrrolic NG are more elongated than those in the pyridinic counterparts. We also observe that, except for the case of Si + Nnit (Fig. 5c, bottom panels), all of them show the bipolar property (Table 2). The different behavior between the Npy family (can be converted to the n-type) and Nnit or Npyrr (can be converted only to the bipolar-type) upon the introduction of a codoping element can be understood from their band structure plots. While the pyridinic defect states can energetically mix with the π states of graphene (Fig. 1b and c, bottom panels), the nitrilic and pyrrolic defect states are energetically located further away from EF and cannot strongly hybridize with the graphene π states (Fig. 1d and e, bottom panels).
In spite of the difference, we can generally conclude that the p-type character of the structurally more distorted and energetically less probable Nnit and Npyrr defects (Table 1) can be also eliminated by B, Al, and P codoping. The COHPs of C–N in these cases are nearly zero around ED, or the N atoms in the codoped Nnit and Npyrr do not affect the sp2 carbon network. The nature of charge distribution in X + Nnit and X + Npyrr can be again visualized in the Mulliken population analysis plot as shown for the representative B + Npyrr case in Fig. 2g, which shows that the net induced charge is localized near the defect sites as in the (bipolar) V2-Npy4 (Fig. 2c).
It remains to be explained why the Si-codoped Nnit maintains the p-type character (Fig. 5c, bottom panel). To understand its origin, we have additionally analyzed the COHPs of various bonds (see ESI Fig. S11†), and found that Si + Nnit can be discriminated from the B/Al/P + Nnit counterparts by a strong antibonding Si–C COHP peak that appears below ED and is pinned near EF (Fig. 5c, bottom panel). On the other hand, the B–C in B + Nnit has bonding character around EF (Fig. 5a, bottom panel).
We finally demonstrate a very desirable feature of the codoping approach in terms of charge transport characteristics. We show in Fig. 6a and b the transmission functions of (codoped) NG for the representative cases of (B+)V1-Npy1 and (P+)V2-Npy1. Note that in these NEGF calculations, unlike the DFT results that have been discussed so far, we are considering isolated doping sites sandwiched by two semi-infinite pristine graphene electrodes (see ESI Fig. S1†). The work function of the entire junction system is thus determined by the pristine graphene region, which results in the Dirac point located at EF. The corresponding current–bias voltage curves calculated according to the Landauer–Büttiker formula,
(6) |
The availability of a simple yet effective method to convert NG with mixed n- and p-type characters into near-uniform n-type NG will improve the performance and reliability of various graphene-based electronic devices such as all-graphene p–n junctions, field-effect transistors, integrated circuits, etc.47–50 We additionally note that the codoping-induced type change in NG may have significant implications for energy applications.10–20 In an effort to identify the catalytically active sites in NG with a significantly improved ORR performance, it was recently shown that the ORR activity in NG-based fuel cell cathodes is proportional to the graphitic N content.17 On the other hand, it was experimentally demonstrated that the codoping of B or P into NG improves the ORR performance.28,29 While the precise origin of the enhanced ORR activity in NG and codoped NG is not yet clear, we can propose that the codoping-induced conversion of mixed-type NG into uniformly n-type NG (i.e. generating more “effectively” graphitic N atoms) and the enhanced charge transport capacity will play an important role in enhancing the ORR activity.
Note added in proof: After the submission of this work for publication, we became aware of ref. 51, in which the authors experimentally synthesized P, N-codoped graphene by a chemical vapor deposition method and found much improved air-stable n-type characteristics. In addition, in ref. 52, the authors theoretically predicted that dissociative adsorption of H2 molecules on the trimerized and tetramerized pyridine-type defects is energetically favorable, and the adsorption of two H atoms changes their electronic properties from p-type to n-type doping.
Footnote |
† Electronic supplementary information (ESI) available: Relative formation energies for the codoping of V1-Npy1–2 and V2-Npy1–3; atomic structures and DOS of F + V1-Npy3 and F + V2-Npy4; atomic structures, DOS, and COHP of the B/Al/Si/P-codoped V1-Npy1–2 and V2-Npy1–3; decomposed COHPs of B/Al/Si/P + V1-Npy3, B/Al/Si/P + V2-Npy4, B + Nnit, and Si + Nnit. See DOI: 10.1039/C4NR05024J |
This journal is © The Royal Society of Chemistry 2014 |