Supporting Information. Theoretical Investigation of the Hydrodeoxygenation of

Transcription

1 Supporting Information Theoretical Investigation of the Hydrodeoxygenation of Levulinic Acid to γ-valerolactone over Ru(0001) Osman Mamun 1, Eric Walker 1, Muhammad Faheem 1,2, Jesse Q. Bond 3, and Andreas Heyden 1,* 1 Department of Chemical Engineering, University of South Carolina, 301 S. Main Street, Columbia, South Carolina 29208, USA 2 Department of Chemical Engineering, University of Engineering & Technology, Lahore 54890, Pakistan. 3 Department of Biomedical and Chemical Engineering, Syracuse University, Syracuse, New York 13244, USA heyden@cec.sc.edu S.1 DFT functional validation 2 S.2 Brønsted-Evans-Polanyi (BEP) and transition state scaling relations for C-O ring opening reactions relevant for the HDO of levulinic acid over Ru(0001) 3 S.3 PBE-D3 optimized structures 6 S.4 Equilibrium and reaction rate constant at various temperatures 11 S.5 Reactant and product partial pressure estimation 15 S.6 Development of microkinetic model 16 S.7 Lateral interaction parameter estimation for hydrogen and oxygen on Ru(0001) 25 S.8 Rates of all elementary reaction (s -1 ) at various reaction temperatures 28 S.9 Surface coverage of all adsorbed surface species at various reaction temperatures 31 S.10 Kinetic degree of rate control of key transition states at 423 K 33 1

2 S.1 DFT functional validation To confirm the accuracy of the PBE-D3 functional, we compare PBE-D3 gas phase reaction energies against CCSD(T) data. In particular, for the coupled cluster calculations, all geometries were first optimized using the MP2/def2-TZVPP level of theory. Then, a coupled cluster (CCSD(T)) single point calculation was performed on the optimized geometry using the same basis set. Both MP2 and CCSD(T) calculations were carried out using the electronic structure program TURBOMOLE 6.0. The zero point energy is computed at the MP2 level of theory. Overall, the deviations of PBE-D3 relative to the coupled cluster calculations are less then 2 kcal/mole. Table S1: Zero point corrected reaction energy for various vapor phase reactions using the PBE-D3 functional and the CCSD(T) level of theory. *experimental value: kcal/mol. 1 2

3 S.2 Brønsted-Evans-Polanyi (BEP) and transition state scaling relations for C-O ring opening reactions relevant for the HDO of levulinic acid over Ru(0001) BEP relations 2-3 have previously been used to predict catalytic phenomena of a similar set of reactions over various catalytic surfaces 4. The Newns-Anderson 5-6 model demonstrated that catalytic action of a transition metal is strongly dependent on the d-band center of that metal. This model explains the fundamental idea of the formulation of BEP relations, i.e., the activation energy, EE AA, of a reaction can be expressed as a linear function of the reaction energy, EE. Similarly, Liu and Hu 7 formulated a relation known as Transition state scaling (TSS) which states that the transition state energy, EE TTTT, can be expressed as a linear function of final state energy, EE FFFF. EE AA = αα EE ββ (1) EE TTTT = αα EE FFFF ββ (2) Considering the difficulty in reliably identifying a transition state for C-O ring opening reactions, we used BEP and TSS relations to predict the activation barriers for these processes from the C-O ring opening reactions for which we could identify transition states with our required accuracy. In particular, we used 12 converged ring opening transition states in our BEP and TSS relations (7 of which are not part of our reaction network). All transition state structures used in the BEP correlations that are not part of our reaction network are shown in Figure S3. C-O ring opening reactions are found to have a BEP slope of 0.08 and an intercept of 0.56 ev with a mean absolute error (MAE) of 0.37 ev. Such a relatively large MAE has also been reported by Wang et al. (0.45 ev) for C-O ring opening reactions. 8 We believe the origin for the relatively large MAE can be found in the large difference in steric orientation of reactants, products and transition 3

4 state structures. For predicting kinetic parameters in our microkinetic model, we used a BEP relation in free energy, which can be represented as GG AA = αα GG ββ (3) and shown in Figure S2. Figure S1: BEP (Left) and TSS (right) relations for C-O bond scisson reaction (ring opening) of various levulinic acid derivatives over Ru (0001) surface Figure S2: BEP relation (free energy of activation as a function of standard free energy of reaction) for C-O ring opening reactions at various temperatures. 4

5 Table S2: BEP estimated free energy of activation at different temperatures for reactions for which no reliable transition state could be identified. Step BBBBBB eeeeeeeeeeeeeeeeee GG (eeee) 323 K 373 K 423 K 523 K 623 K R R R R

6 S.3 PBE-D3 optimized structures LA Al Hy HPA I-01 I-02 I-03 I-04 I-05 I-06 I-07 I-08 6

7 I-09 I-10 I-11 I-12 I-13 I-14 I-15 I-16 I-17 I-18 GVL Figure S3: PBE-D3 optimized geometries of various surface intermediates, reactants, and products. 7

8 TS-3 TS-4 TS-5 TS-6 TS-7 TS-8 TS-9 TS-10 TS-11 TS-12 TS-13 TS-14 TS-15 TS-16 TS-17 TS-18 TS-21 TS-22 8

9 TS-23 TS-24 TS-25 TS-26 TS-27 TS-28 TS-29 TS-30 TS-31 TS-32 TS-34 TS-35 TS-36 TS-37 TS-38 Figure S4: PBE-D3 optimized transition state structures. The transition state number corresponds to the elementary reaction number in Table 2. 9

10 Figure S5: To increase the number of data points in our BEP relations for C-O ring opening reactions, we added 7 transition state structures (shown above) that are not part of the reaction network to the BEP relation calculation. 10

11 S.4 Equilibrium and reaction rate constant at various temperatures Table S3: Equilibrium and reaction rate constants for all elementary reaction steps considered in the development of the microkinetic model. *All rate constants are in units of inverse seconds and bar. Temperature (K) Constant Step rr 11 KK kk rr 22 KK kk rr 33 KK kk rr 44 KK kk rr 55 KK kk rr 66 KK kk rr 77 KK kk rr 88 KK kk rr 99 KK kk rr 1111 KK kk rr 1111 KK

12 kk rr 1111 KK kk rr 1111 KK kk rr 1111 KK kk rr 1111 KK kk rr 1111 KK kk rr 1111 KK kk rr 1111 KK kk rr 1111 KK kk rr 2222 KK kk rr 2222 KK kk rr 2222 KK kk rr 2222 KK kk rr 2222 KK kk rr 2222 KK kk

13 rr 2222 KK kk rr 2222 KK kk rr 2222 KK kk rr 2222 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 3333 KK kk rr 4444 KK

14 kk rr 4444 KK kk rr 4444 KK kk rr 4444 KK kk

15 S.5 Reactant and product partial pressure estimation To compare our modeling results with experimental observations, we used typical reaction conditions in our models. Specifically, experiments are often performed under liquid phase conditions with a concentration of LA ( M), GVL ( tttt MM) and hydrogen (10 bar) 9. As a result, we used for our microkinetic model a partial pressure of 10 bar for hydrogen. To find the corresponding partial pressure/fugacity of LA for a 0.45 M solution, we used the modified Raoult s law, ff LLLL vv ssssss = PP LLLL = xx LLLL γγ LLLL PP LLLL Here, thermodynamic data such as the activity coefficient and saturation pressure of LA were calculated using the COSMOtherm program package To compute the mole fraction of LA in solution for a 0.45 M solution, we used xx LLLL = 0.45 MM vv mm3 kkkk 18 kkkk kkkkkkkk (i.e., we assumed a dilute solution). Here, vv is the specific volume of saturated, liquid water, for which we used standard thermodynamic property data 13. Table S4 summarizes all data used to calculate the fugacity/partial pressure of LA at various temperatures. Since the LA to GVL ratio is about 900 under experimental reaction conditions, we assumed a conversion of 0.1% to estimate the fugacity of GVL. Table S4: LA partial pressure estimation using modified Roult s law 15

16 S.6 Development of microkinetic model In the following, we define the number of intermediate species i on the surface over the total number of surface sites (Ru(0001) surface atoms), θθ ii as, θθ ii = NNNNNNNNNNNN oooo iiiiiiiiiiiiiiiiiiiiiiii ii ssssssssssssss aaaaaaaaaaaaaaaa oooo tthee ssssssssssssss TTTTTTTTTT nnnnnnnnnnnn oooo aaaaaaaaaaaa ssssssssssssss ssssssssss We distinguish θθ ii from the surface coverage, θθ ii, which is the product of θθ ii and the number of sites occupied by intermediate i (please see the main article for a description of each surface intermediate I-i and rr ii ). We note that θθ ii is the activity of a lattice gas that occupies multiple sites. 14 As a result, we define the following set of elementary reactions: rr 1 = kk 1ff PP LLLL θθ 2 kk 1bb θθ LLLL rr 2 = kk 2ff PP HH2 θθ 2 kk 2bb θθ HH 2 rr 3 = kk 3ff θθ LLLL θθ HH kk 3bb θθ AAAA θθ rr 4 = kk 4ff θθ LLLL θθ HH kk 4bb θθ HHHH θθ rr 5 = kk 5ff θθ AAAA θθ HH kk 5bb θθ HHHHHH rr 6 = kk 6ff θθ HHHH θθ HH kk 6bb θθ HHHHHH rr 7 = kk 7ff θθ HHHHHH θθ kk 7bb θθ 01 θθ OOOO rr 8 = kk 8ff θθ HHHHHH θθ HH kk 8bb θθ 02 θθ 2 rr 9 = kk 9ff θθ 17 θθ HH kk 9bb θθ 18 θθ 2 rr 10 = kk 10ff θθ HHHHHH kk 10bb θθ 03 θθ OOOO rr 11 = kk 11ff θθ 01 θθ kk 11bb θθ 06 θθ HH rr 12 = kk 12ff θθ 01 θθ HH kk 12bb θθ 04 θθ 3 rr 13 = kk 13ff θθ 02 kk 13bb θθ 04 θθ OOOO 16

17 rr 14 = kk 14ff θθ AAAA θθ 2 kk 14bb θθ 06 θθ OOOO rr 15 = kk 15ff θθ 02 θθ kk 15bb θθ 05 θθ HH rr 16 = kk 16ff θθ AAAA θθ HH kk 16bb θθ 05 θθ rr 17 = kk 17ff θθ AAAA kk 17bb θθ 18 θθ rr 18 = kk 18ff θθ HHHH kk 18bb θθ 14 θθ OOOO rr 19 = kk 19ff θθ HHHH kk 19bb θθ 13 θθ rr 20 = kk 20ff θθ 03 kk 20bb θθ 08 rr 21 = kk 21ff θθ 03 θθ 2 kk 21bb θθ 09 θθ HH rr 22 = kk 22ff θθ 04 θθ 2 kk 22bb θθ 07 θθ HH rr 23 = kk 23ff θθ 05 θθ kk 23bb θθ 07 θθ OOOO rr 24 = kk 24ff θθ 07 kk 24bb θθ 08 rr 25 = kk 25ff θθ 18 θθ 2 kk 25bb θθ GGGGGG θθ OOOO rr 26 = kk 26ff θθ 14 θθ kk 26bb θθ 16 rr 27 = kk 27ff θθ 13 θθ 2 kk 27bb θθ 16 θθ OOOO rr 28 = kk 28ff θθ 16 kk 28bb θθ 15 θθ HH rr 29 = kk 29ff θθ 16 θθ HH kk 29bb θθ 08 θθ rr 30 = kk 30ff θθ 06 kk 30bb θθ GGGGGG θθ rr 31 = kk 31ff θθ 08 θθ kk 31bb θθ GGGGGG θθ HH rr 32 = kk 32ff θθ 15 θθ HH kk 32bb θθ GGGGGG rr 33 = kk 33ff θθ 09 kk 33bb θθ GGGGGG θθ rr 34 = kk 34ff θθ LLLL θθ 2 kk 34bb θθ 10 θθ OOOO rr 35 = kk 35ff θθ LLLL θθ 2 kk 35bb θθ 11 θθ HH rr 36 = kk 36ff θθ 10 kk 36bb θθ 15 θθ 2 17

18 rr 37 = kk 37ff θθ 10 θθ HH kk 37bb θθ 06 θθ rr 38 = kk 38ff θθ 11 θθ HH kk 38bb θθ 12 θθ rr 39 = kk 39ff θθ 12 kk 39bb θθ 17 θθ rr 40 = kk 40ff θθ GGGGGG kk 40bb PP GGGGGG θθ 2 rr 41 = kk 41ff θθ HH θθ OOOO kk 41bb θθ HH2 OOθθ ΔΔ rr 42 = kk 42ff θθ HH2 OO kk 42bb θθ PP HH2 OO rr 43 = kk 43ff θθ OOOO θθ kk 43bb θθ HH θθ OO rr 44 = kk 44ff θθ HHHHHH kk 44bb PP HHHHHH θθ 3 Species balance equations for all surface species and the overall mass balance lead to: ddθθ LLLL = rr 1 rr 3 rr 4 rr 34 rr 35 ddθθ AAAA = rr 3 rr 5 rr 14 rr 16 rr 17 ddθθ HHHHHH ddθθ HHHH ddθθ 06 = rr 4 rr 6 rr 18 rr 19 = rr 5 rr 6 rr 7 rr 8 rr 10 rr 44 ddθθ 01 ddθθ 02 ddθθ 03 ddθθ 04 ddθθ 05 = rr 7 rr 11 rr 12 = rr 8 rr 13 rr 15 = rr 10 rr 20 rr 21 = rr 12 rr 13 rr 22 = rr 15 rr 16 rr 23 = rr 11 rr 14 rr 30 rr 37 18

19 ddθθ GGGGGG ddθθ 08 ddθθ 16 ddθθ 07 ddθθ 10 ddθθ 15 = rr 22 rr 23 rr 24 = rr 20 rr 24 rr 31 rr 29 ddθθ 09 ddθθ 11 ddθθ 12 ddθθ 13 ddθθ 14 = rr 21 rr 33 = rr 34 rr 36 rr 37 = rr 35 rr 38 = rr 38 rr 39 = rr 19 rr 27 = rr 18 rr 26 = rr 28 rr 32 rr 36 = rr 26 rr 27 rr 28 rr 29 ddθθ 18 ddθθ 17 = rr 39 rr 9 = rr 9 rr 17 rr 25 = rr 25 rr 30 rr 31 rr 32 rr 33 rr 40 ddθθ HH = 2rr 2 rr 3 rr 4 rr 5 rr 6 rr 8 rr 9 rr 11 rr 12 rr 15 rr 16 rr 21 rr 22 rr 28 rr 29 rr 31 rr 32 rr 35 rr 37 rr 38 rr 41 rr 43 ddθθ OOOO = rr 7 rr 10 rr 13 rr 14 rr 18 rr 23 rr 25 rr 27 rr 34 rr 41 rr 43 ddθθ HH2 OO = rr 41 rr 42 19

20 ddθθ OO = rr 43 And 2θθ LLLL 3θθ HHHHHH 2θθ AAAA 2θθ HHHH 3θθ 01 2θθ 02 2θθ 03 θθ 04 2θθ 05 3θθ 06 2θθ 07 2θθ 08 3θθ 09 3θθ 10 3θθ 11 3θθ 12 θθ 13 θθ 14 θθ 15 2θθ 16 2θθ 17 θθ 18 2θθ GGGGGG θθ HH θθ OOOO θθ OO θθ HH2 OO θθ = 1.00 For the 2-site model, we used the following set of elementary reactions and differential surface species balances. rr 1 = kk 1ff PP LLLL θθ 2 kk 1bb θθ LLLL rr 2 = kk 2ff PP HH2 θθ ΔΔ 2 kk 2bb θθ HH 2 rr 3 = kk 3ff θθ LLLL θθ HH kk 3bb θθ AAAA θθ ΔΔ rr 4 = kk 4ff θθ LLLL θθ HH kk 4bb θθ HHHH θθ ΔΔ rr 5 = kk 5ff θθ AAAA θθ HH θθ kk 5bb θθ HHHHHH θθ ΔΔ rr 6 = kk 6ff θθ HHHH θθ HH θθ kk 6bb θθ HHHHHH θθ ΔΔ rr 7 = kk 7ff θθ HHHHHH θθ kk 7bb θθ 01 θθ OOOO rr 8 = kk 8ff θθ HHHHHH θθ HH kk 8bb θθ 02 θθ θθ ΔΔ rr 9 = kk 9ff θθ 17 θθ HH kk 9bb θθ 18 θθ θθ ΔΔ rr 10 = kk 10ff θθ HHHHHH kk 10bb θθ 03 θθ OOOO rr 11 = kk 11ff θθ 01 θθ ΔΔ kk 11bb θθ 06 θθ HH rr 12 = kk 12ff θθ 01 θθ HH kk 12bb θθ 04 θθ 2 θθ ΔΔ rr 13 = kk 13ff θθ 02 kk 13bb θθ 04 θθ OOOO rr 14 = kk 14ff θθ AAAA θθ 2 kk 14bb θθ 06 θθ OOOO rr 15 = kk 15ff θθ 02 θθ ΔΔ kk 15bb θθ 05 θθ HH rr 16 = kk 16ff θθ AAAA θθ HH kk 16bb θθ 05 θθ ΔΔ 20

21 rr 17 = kk 17ff θθ AAAA kk 17bb θθ 18 θθ rr 18 = kk 18ff θθ HHHH kk 18bb θθ 14 θθ OOOO rr 19 = kk 19ff θθ HHHH kk 19bb θθ 13 θθ rr 20 = kk 20ff θθ 03 kk 20bb θθ 08 rr 21 = kk 21ff θθ 03 θθ θθ ΔΔ kk 21bb θθ 09 θθ HH rr 22 = kk 22ff θθ 04 θθ θθ ΔΔ kk 22bb θθ 07 θθ HH rr 23 = kk 23ff θθ 05 θθ kk 23bb θθ 07 θθ OOOO rr 24 = kk 24ff θθ 07 kk 24bb θθ 08 rr 25 = kk 25ff θθ 18 θθ 2 kk 25bb θθ GGGGGG θθ OOOO rr 26 = kk 26ff θθ 14 θθ kk 26bb θθ 16 rr 27 = kk 27ff θθ 13 θθ 2 kk 27bb θθ 16 θθ OOOO rr 28 = kk 28ff θθ 16 θθ ΔΔ kk 28bb θθ 15 θθ HH θθ rr 29 = kk 29ff θθ 16 θθ HH kk 29bb θθ 08 θθ ΔΔ rr 30 = kk 30ff θθ 06 kk 30bb θθ GGGGGG θθ rr 31 = kk 31ff θθ 08 θθ ΔΔ kk 31bb θθ GGGGGG θθ HH rr 32 = kk 32ff θθ 15 θθ HH θθ kk 32bb θθ GGGGGG θθ ΔΔ rr 33 = kk 33ff θθ 09 kk 33bb θθ GGGGGG θθ rr 34 = kk 34ff θθ LLLL θθ 2 kk 34bb θθ 10 θθ OOOO rr 35 = kk 35ff θθ LLLL θθ θθ ΔΔ kk 35bb θθ 11 θθ HH rr 36 = kk 36ff θθ 10 kk 36bb θθ 15 θθ 2 rr 37 = kk 37ff θθ 10 θθ HH kk 37bb θθ 06 θθ ΔΔ rr 38 = kk 38ff θθ 11 θθ HH kk 38bb θθ 12 θθ ΔΔ rr 39 = kk 39ff θθ 12 kk 39bb θθ 17 θθ 21

22 rr 40 = kk 40ff θθ GGGGGG kk 40bb PP GGGGGG θθ 2 rr 41 = kk 41ff θθ HH θθ OOOO kk 41bb θθ HH2 OOθθ ΔΔ rr 42 = kk 42ff θθ HH2 OO kk 42bb θθ PP HH2 OO rr 43 = kk 43ff θθ OOOO θθ ΔΔ kk 43bb θθ HH θθ OO rr 44 = kk 44ff θθ HHHHHH kk 44bb PP HHHHHH θθ 3 Species balance equations for all surface species and the overall mass balance lead to: ddθθ LLLL = rr 1 rr 3 rr 4 rr 34 rr 35 ddθθ AAAA = rr 3 rr 5 rr 14 rr 16 rr 17 ddθθ HHHHHH ddθθ HHHH ddθθ 06 ddθθ 08 = rr 4 rr 6 rr 18 rr 19 = rr 5 rr 6 rr 7 rr 8 rr 10 rr 44 ddθθ 01 ddθθ 02 ddθθ 03 ddθθ 04 ddθθ 05 ddθθ 07 = rr 7 rr 11 rr 12 = rr 8 rr 13 rr 15 = rr 10 rr 20 rr 21 = rr 12 rr 13 rr 22 = rr 15 rr 16 rr 23 = rr 11 rr 14 rr 30 rr 37 = rr 22 rr 23 rr 24 = rr 20 rr 24 rr 31 rr 29 22

23 ddθθ GGGGGG ddθθ 16 ddθθ 10 ddθθ 15 ddθθ 09 ddθθ 11 ddθθ 12 ddθθ 13 ddθθ 14 = rr 21 rr 33 = rr 34 rr 36 rr 37 = rr 35 rr 38 = rr 38 rr 39 = rr 19 rr 27 = rr 18 rr 26 = rr 28 rr 32 rr 36 = rr 26 rr 27 rr 28 rr 29 ddθθ 18 ddθθ 17 = rr 39 rr 9 = rr 9 rr 17 rr 25 = rr 25 rr 30 rr 31 rr 32 rr 33 rr 40 ddθθ HH = 2rr 2 rr 3 rr 4 rr 5 rr 6 rr 8 rr 9 rr 11 rr 12 rr 15 rr 16 rr 21 rr 22 rr 28 rr 29 rr 31 rr 32 rr 35 rr 37 rr 38 rr 41 rr 43 ddθθ OOOO = rr 7 rr 10 rr 13 rr 14 rr 18 rr 23 rr 25 rr 27 rr 34 rr 41 rr 43 ddθθ HH2 OO = rr 41 rr 42 ddθθ OO = rr 43 23

24 2θθ LLLL 3θθ HHHHHH 2θθ AAAA 2θθ HHHH 3θθ 01 2θθ 02 2θθ 03 θθ 04 2θθ 05 3θθ 06 2θθ 07 2θθ 08 3θθ 09 3θθ 10 3θθ 11 3θθ 12 θθ 13 θθ 14 θθ 15 2θθ 16 2θθ 17 θθ 18 2θθ GGGGGG θθ OOOO θθ OO θθ HH2 OO θθ = 1.00 and θθ HH θθ ΔΔ =

25 S.7 Lateral interaction parameter estimation for hydrogen and oxygen on Ru(0001) Table S5: Hydrogen lateral interaction estimation data Coverage Average energy (ev) Integral energy (DFT) Fitted integral energy Table S6: Oxygen lateral interaction estimation data Coverage Average energy (ev) Integral energy (DFT) Fitted integral energy

26 EE aaaaaa = (EE mmmmmmmmmmaaaaaaaaaaaaaaaaaa EE ssssssss nn 2 EE HH 2 ) nn where nn is the number of adsorbed hydrogen atoms. The integral energy (DFT) is calculated as EE iiiiiiiiiiiiiiii DDDDDD = EE aaaaaa θθ HH and the fitted integral energy is calculated as (where EE 0 is the low coverage adsorption energy and εε and θθ HH,0 are fitting parameters) 15 EE ffffffffffff DDDDDD = EE 0 θθ HH εε(θθ HH θθ HH,0 ) 2 EEEEEEEEEE = EE iiiiiiiiiiiiiiii ffffffffffff DDDDDD EE DDDDDD The sum of the square of the error was minimized to find the parameters of this two parameter model (EE 0 is taken to be equal to the adsorption energy at 1/16 ML). (A) 26

27 (B) Figure S6: Coverage dependent adsorption energy calculated using a two parameter lateral interaction model for (A) hydrogen and (B) oxygen. 27

28 S.8 Rates of all elementary reaction (s -1 ) at various reaction temperatures Table S7: Rates of all the elementary reaction rates at various temperatures for the one site model Reaction step Temperature (K) r1 2.67E E E E E-01 r2 2.67E E E E E-01 r3 2.67E E E E E-01 r4 1.04E E E E E-07 r5 1.02E E E E E-07 r6 9.82E E E E E-07 r7 2.00E E E E E-07 r8 7.48E E E E E-13 r9-2.55e e e e E-10 r E E E E E-12 r E E E E E-07 r E E E E E-10 r E E E E E-11 r E E E E E-03 r e e e e e-11 r E E E E E-11 r E E E E E-01 r E E E E E-08 r E E E E E-12 r E E E E E-06 r e e e e e-06 r E E E E E-10 r E E E E E-13 r E E E E E-10 r E E E E E-01 r E E E E E-08 r E E E E E-12 r E E E E E-06 r e e e e e-06 r E E E E E-03 r e e e e e-11 r E E E E E-03 r e e e e e-06 28

29 Reaction step Temperature (K) r E E E E E-07 r e e e e E-10 r E E E E E-03 r e e e e e-03 r e e e e E-10 r e e e e E-10 r E E E E E-01 r E E E E E-01 r E E E E E-01 r e E E E E-15 rr E E E E E-08 Table S8: Rates of all the elementary reaction rates at various temperatures for the two site model Reaction step Temperature (K) r1 1.94E E E E E00 r2 1.94E E E E E00 r3 1.94E E E E E00 r4 4.90E E E E E-07 r5 9.40E E E E E-09 r6 3.55E E E E E-09 r7-1.43e e e e e-09 r8 2.41E E E E E-13 r9-8.11e e e e e-09 r E E E E E-12 r e e e e e-09 r E E E E E-10 r E E E E E-11 r E E E E E-02 r e e e e e-11 r E E E E E-11 r E E E E E00 r E E E E E-07 r E E E E E-11 r E E E E E-08 r e e e e e-08 r E E E E E-10 r E E E E E-13 r E E E E E-10 r E E E E E00 29

30 Reaction Temperature (K) step r E E E E E-07 r E E E E E-11 r E E E E E-07 r e e e e e-08 r E E E E E-02 r e e e e e-13 r E E E E E-05 r e e e e e-08 r E E E E E-06 r e e e e e-09 r e e E E E-05 r E E E E E-05 r e e e e e-09 r e e e e e-09 r E E E E E00 r E E E E E00 r E E E E E00 r e e E E E-15 rr E E E E E-08 30

31 S.9 Surface coverage of all adsorbed surface species at various reaction temperatures Table S9: Surface coverage of all surface species at various temperatures for the one site model Adsorbed Temperature (K) species I E E E E E-09 I E E E E E-16 I E E E E E-10 I E E E E E-09 I E E E E E-15 I E E E E E-04 I E E E E E-11 I E E E E E-14 I E E E E E-07 I E E E E E-08 I E E E E E-02 I E E E E E-06 I E E E E E-11 I E E E E E-06 I E E E E E-07 I E E E E E-12 I E E E E E-14 I E E E E E-07 LA 2.25E E E E E-07 Al 1.25E E E E E-15 Hy 1.68E E E E E-08 HPA 1.08E E E E E-13 H 1.00E E E E E-01 OH 2.17E E E E E-04 GVL 6.56E E E E E-09 Free sites 1.47E E E E E-03 H2O 1.45E E E E E-09 O 1.94E E E E E-03 Note: All reported numbers are the surface coverage over the number of adsorption sites. 31

32 Table S10: Surface coverage of all the surface species at various temperatures for the two site model Adsorbed Temperature (K) species I E E E E E-09 I E E E E E-17 I E E E E E-10 I E E E E E-09 I E E E E E-15 I E E E E E-03 I E E E E E-10 I E E E E E-13 I E E E E E-06 I E E E E E-07 I E E E E E-01 I E E E E E-06 I E E E E E-11 I E E E E E-08 I E E E E E-06 I E E E E E-13 I E E E E E-13 I E E E E E-07 LA 1.64E E E E E-06 Al 4.25E E E E E-16 Hy 2.07E E E E E-08 HPA 9.88E E E E E-12 H 1.00E E E E E-01 OH 1.57E E E E E-04 GVL 4.77E E E E E-09 Free sites (*) 3.97E E E E E-03 H2O 3.96E E E E E-09 O 2.12E E E E E-04 Free sites (Δ) 1.48E E E E E-03 32

33 S.10 Kinetic degree of rate control of key transition states at 423 K Table S11: Kinetic rate control of key reaction steps for both the one- and two-site model Step KKKKKK (1 site microkinetic model) KKKKKK (2 site microkinetic model) rr E E-01 rr E E-01 rr E E-03 rr E E-03 rr E E-03 Bibliography 1. Serrano-Ruiz, J. C.; West, R. M.; Dumesic, J. A., Annu. Rev. Chem. Biomol. Eng. 2010, 1, Bronsted, J. N., Chem. Rev. 1928, 5, Evans, M. G.; Polanyi, M., Trans. Faraday Soc. 1938, 34, Stamatakis, M.; Vlachos, D. G., ACS Catal. 2012, 2, Newns, D. M., Phys. Rev. 1969, 178, Anderson, P. W., Phys. Rev. 1961, 124, Liu, Z. H.; Hu, P.; Liu, Z. H.; Hu, P., J. Chem. Phys. 2001, 115, Wang, S.; Vorotnikov, V.; Sutton, J. E.; Vlachos, D. G., ACS Catal. 2014, 4, Abdelrahman, O. A.; Heyden, A.; Bond, J. Q., ACS Catal. 2014, 4, Schafer, A.; Klamt, A.; Sattel, D.; Lohrenz, J. C. W.; Eckert, F., Phys. Chem. Chem. Phys. 2000, 2, Ahlrichs, R.; Bär, M.; Häser, M.; Horn, H.; Kölmel, C., Chem. Phys. Lett. 1989, 162, Treutler, O.; Ahlrichs, R., J. Chem. Phys. 1995, 102, Elliott, J. R.; Lira, C. T., Introductory Chemical Engineering Thermodynamics. Prentice Hall PTR: Nitta, T.; Shigetomi, T.; Kuro-Oka, M.; Katayama, T., J. Chem. Eng. Jpn. 1984, 17, Grabow, L.; Hvolbæk, B.; Nørskov, J., Top. Catal. 2010, 53,