Drivers and barriers to the diffusion of energy-efficient technologies—a plant-level analysis of the German steel industry

Abstract

The paper aims at explaining why large-scale energy-intensive industries—here the German iron and steel industry—had a period of slow uptake of major energy-efficient technologies from the mid 1990s to mid 2000s (Arens and Worrell, 2014) and why from the mid 2000s onwards these technologies are increasingly implemented again. We analyze the underlying factors and investment/innovation behavior of individual firms in the German iron and steel industry to better understand barriers and drivers for technological change. The paper gives insights on the decision-making process on energy efficiency in firms and helps to understand how policy affects decision-making. We use a mixed method approach. First, we analyze the diffusion of three energy-efficient technologies (EET) for primary steelmaking from their introduction until today (top-pressure recovery turbine (TRT), basic oxygen furnace gas recovery (BOFGR), and pulverized coal injection (PCI)). We derive the uptake of these technologies both at the national level and at the level of the individual firm. Second, we analyze the impact of drivers and barriers on the decision-making process of individual firms whether or not they want to implement these technologies. Economics and access to capital are the foremost barriers to the uptake of an EET. If the expected payback period exceeds a certain value or if the company lacks capital, investments in EET seem not to happen. But even if an EET is economically viable and the company has access to capital, investments in EET might not be realized. Policy-induced prices might have strengthened the recent diffusion of TRT. We found indications that in a limited number of cases, policy intervention was a driving factor. Technical risks and imperfect information are only marginal factors in our cases. Site-specific factors seem to be important, as site-specific factors shape the economics of the selected EET.

Appendix

Data sources for the construction of the timeline of energy prices for the German industry

Table 8 Data sources for the construction of the timeline of energy prices for the German industry

Calculation of payback periods

The payback periods are calculated according to Eq. (2).

$$ {\mathrm{PB}}_{i,t}(k)=\frac{I_i}{P_{i,t}(k)-{C}_i(k)} $$

(2)

where

PB: payback period

I: investment

P: annual return due to technology i

C: annual O&M cost

i: technology

t: case (Table 4)

k: year

The proceeds for TRT are assumed to equal the market value of generated electricity. This assumes that the company purchases electricity from the public grid and that by applying TRT the company can reduce its electricity consumption from the public grid. Later, we will see that not all steel companies in Germany fulfill this assumption and that TRT in some cases does not compete with the electricity price from the public grid but with the price for on-site electricity generation. Hence, the electricity prices for cases (b) and (c) are (Eq. (3)–(4)):

$$ \left(\mathrm{b}\right){PC}_{\mathrm{EL},b}(k)={PC}_{\mathrm{EL},a}(k)-{SE}_{CO2,\mathrm{EL}}(k)\cdot {PC}_{CO2}(k), $$

(3)

$$ \left(\mathrm{c}\right){PC}_{\mathrm{EL},c}(k)={PC}_{\mathrm{EL},a}(k)-{SE}_{CO2,\mathrm{EL}}(k)\cdot {PC}_{CO2}(k)-EEG(k). $$

(4)

where

PC: price

EL: electricity

SE: specific emissions

CO₂: carbon dioxide

EEG: levy for the support of renewable energies

k: year

a, b: case (Table 4).

For BOFGR, we assume that the recovered BOFG reduces the consumption of natural gas and that in case of EU-ETS costs for the CO₂ emissions of that amount of natural gas are saved.

Estimating the economic benefits of PCI, we follow the approach of Schott et al. (2012). Coke is partly replaced by coal in the blast furnace. Differences between the coke and coal price lead to economic benefits. CO₂ emission reductions are calculated by accounting coal consumption both in coke ovens and blast furnaces. We assume that 1 t of coke is produced from 1.3 t of coal (Schott et al. 2012). Table 9 lists further assumptions.

Table 9 Assumptions on the selected energy-efficient technologies (EETs)

The calculations of the proceeds of the respective technologies are given below (Eq. (5)–9).

$$ {P}_{TRT,t}={SP}_{\mathrm{EL},TRT}\cdot {PC}_{\mathrm{EL},t}(k)\cdot {CP}_{BF} $$

(5)

$$ {P}_{\mathrm{BOFGR},\mathrm{a}}={SR}_{\mathrm{BOFGR}}\cdot {PC}_{NG}(k)\cdot {CP}_{BOF} $$

(6)

$$ {P}_{\mathrm{BOFG}\mathrm{R},b}={SR}_{\mathrm{BOFG}}\cdot \left({PC}_{NG}(k)+{SE}_{CO2,NG}\cdot {PC}_{CO2}(k)\right)\cdot {CP}_{BOF} $$

(7)

$$ {P}_{PCI,\mathrm{a}}=\left({SC}_{CK,\mathrm{OP}}\cdot {PC}_{CK}(k)-\left({SC}_{CK,WP}\cdot {PC}_{CK}(k)+{SC}_{CL,WP}\cdot {PC}_{CL}(k)\right)\right)\cdot {CP}_{BF} $$

(8)

$$ {P}_{PCI,\mathrm{b}}={P}_{PCI,\mathrm{a}}+\left({SC}_{CK,\mathrm{OP}}\cdot Q-\left({SC}_{CK,WP}\cdot Q+{SC}_{CL,WP}\right)\cdot {SE}_{CO2,CL} \cdot {PC}_{CO2}(k)\right)\cdot {CP}_{BF} $$

(9)

where

P _i: annual return due to technology i

SP: specific production

SR: specific recovery

SE: specific emissions

SC: specific consumption

PC: price

CP: capacity

Q: factor (input to coke oven: coke/coal = 1.6 (Schott et al. 2012)

BF: blast furnace

BOF: basic oxygen furnace

BOFG: basic oxygen furnace gas

CK: coke

CL: coal

CO₂: carbon dioxide

EL: electricity

NG: natural gas

OP: without PCI

WP: with PCI

k: year

t: case a, b, or c (Table 4)

a, b, c: cases (Table 4)