BioEnergy Research

https://doi.org/10.1007/s12155-021-10272-6

Genotype-Environment-Management Interactions in Biomass Yield

and Feedstock Composition of Photoperiod-Sensitive
Energy Sorghum
August Schetter 1,2 & Cheng-Hsien Lin 1 & Colleen Zumpf 1 & Chunhwa Jang 1,2 &
Leo Hoffmann Jr 3 & William Rooney 2,3 & D. K. Lee 1,2,4

Received: 16 October 2020 / Accepted: 17 March 2021

# The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021

Abstract
Recently introduced photoperiod-sensitive (PS) biomass sorghum (Sorghum bicolor L. Moench) needs to be investigated for
their yield potential under different cultivation environments with reasonable nitrogen (N) inputs. The objectives of this study
were to (1) evaluate the biomass yield and feedstock quality of four sorghum hybrids with different levels of PS ranging from
very PS (VPS) hybrids and to moderate PS (MPS) hybrids, and (2) determine the optimal N inputs (0~168 kg N ha−1) under four
environments: combinations of both temperate (Urbana, IL) and subtropical (College Station, TX) regions during 2018 and 2019.
Compared to TX, the PS sorghums in central IL showed higher yield potential and steady feedstock production with an extended
day length and with less precipitation variability, especially for the VPS hybrids. The mean dry matter (DM) yields of VPS
hybrids were 20.5 Mg DM ha−1 and 17.7 Mg DM ha−1 in IL and TX, respectively. The highest N use efficiency occurred at a low
N rate of 56 kg N ha−1 by improving approximately 33 kg DM ha−1 per 1.0 kg N ha−1 input. Approximately 70% of the PS
sorghum biomass can be utilized for biofuel production, consisting of 58-65% of the cell-wall components and 4-11% of the
soluble sugar. This study demonstrated that the rainfed temperate area (e.g., IL) has a great potential for the sustainable cultivation
of PS energy sorghum due to their observed high yield potential, stable production, and low N requirements.

Keywords Photoperiod-sensitive sorghum . Environment . Feedstock quality . Nitrogen use efficiency

Introduction biofuel (e.g., ethanol). Based on its types and primary crop
components, sorghum can be divided into several categories:
Herbaceous feedstock is a critical renewable resource for grain sorghum with starch-rich grain, sweet sorghum with
bioenergy production necessary to alleviate global depen- high soluble sugar, and forage sorghum with high leaf to stem
dence on fossil fuels [1]. Sorghum (Sorghum bicolor L. ratios, silage sorghums and biomass sorghum (bioenergy)
Moench), primarily known as a cereal or forage crop, is re- with high lignocellulosic components [2, 3].
ceiving increased attention as an energy crop for producing Biomass and sweet sorghum are considered an ideal
bioenergy crop for several reasons. Both types have excellent
biomass yield potential, including energy-rich compositions
August Schetter and Cheng-Hsien Lin contributed equally to this work. such as structural and non-structural carbohydrates. In gener-
al, sorghum also has high tolerance to environmental stress
* D. K. Lee
(e.g., drought conditions, high salinity soils, or low soil fertil-
[email protected]
ity) and has the potential to sustain relatively high biomass
1
Department of Crop Sciences, University of Illinois at production under adverse environments. Additionally, the
Urbana-Champaign, Urbana, IL 61801, USA long history of established sorghum production systems in
2
Center for Advanced Bioenergy and Bioproducts Innovation, the USA provides growers with the opportunity to quickly
University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA adopt a new crop into their existing production systems.
3
Department of Soil and Crop Sciences, Texas A&M University, Finally, the genetic improvement of sorghum through tradi-
College Station, TX 77843, USA tional and genomic approaches for multiple objectives has
4
Carl R. Woese Institute for Genomic Biology, University of Illinois allowed for advantageous gains in yield potential, stress toler-
at Urbana-Champaign, Urbana, IL 61801, USA ance, and feedstock quality [3, 4].
 Bioenerg. Res.

Plant breeders have utilized the knowledge of photoperiodism over 5 years (2008-2012). The aboveground dry biomass likely
as a strategy to adjust sorghum maturity to maximize the vege- increased by improving management practices, such as adequate
tative growth phase to improve biomass yield. These manifest as nitrogen (N) input and irrigation systems [7, 10, 20, 24]. Also,
photoperiod-sensitive (PS) sorghum hybrids [4, 5]. The PS sor- environmental variability substantially influenced biomass com-
ghum hybrids remain in the vegetative growth phase until day position. Packer [25] examined the compositional variability in
length is less than a specific day length, which varies depending 15 PS sorghum hybrids in five environments based on different
on the genetics of the hybrid. As such, in temperate environ- soil types and climate regions across TX and showed that the
ments, grain or seed production is essentially impossible because concentrations of cellulose ranged from 26.9 to 31.8%, xylan
day length during the growing season (April-September) is much from 14.9 to 18.4%, and lignin from 8.3 to 18.9%. For the field
greater than the minimum required to initiate reproductive management effects, McKinley et al. [12] reported that irrigation
growth [4, 6]. An extended period of vegetative growth provides increased the concentration of the overall cell-wall composition
sorghum with two significant advantages including (1) extending of the PS hybrids by 20% compared to the non-irrigated fields.
the duration of utilizing the solar radiation to cumulate biomass Although Amaducci et al. [20] did not observe a significant
during the growing season, and (2) a high level of drought toler- management effect, including nutrient input and tillage practices,
ance in rainfed cultivation environment [4, 7, 8]. For instance, the on biomass composition, Almodares et al. [26] showed that the
delayed maturity of the PS energy sorghum is usually associated increased N input reduced both soluble carbohydrates and fiber
with higher dry biomass yield ranging from 20 to 35 Mg ha−1 content.
compared to grain sorghum (approximately twofold higher) [7, Many studies indicated that N application is the predomi-
9, 10]. Under good growing conditions (e.g., without water and nant energy input for lignocellulosic energy crops among oth-
nutrient deficiency), substantially increased biomass yields (e.g., er management operations [27–29]. Compared to grain sor-
>35 Mg DM ha−1) were also reported in TX [11–13]. ghum and other biofuel feedstock resources (e.g., sugarcane
In addition to high yield potential, biomass feedstock compo- and maize), biomass sorghum requires less N input to accu-
sition is also a crucial factor for biofuel conversion processes. mulate lignocellulosic components. The lower N requirement
Comprehensive understanding of the feedstock composition is implies that biomass sorghum can minimize the production
important for multiple operational purposes, including storage, cost and the environmental impact while maximizing biomass
pre-treatments, and biorefinery operations [14, 15]. Different sor- yield [26, 30]. Olsen et al. [31] showed that the PS biomass
ghum types and hybrids contain specific compositional charac- energy sorghum had higher N use efficiency (NUE) than other
teristics. For instance, the PS biomass sorghum hybrids tend to candidate bioenergy species (e.g., grain and sweet sorghum,
have concentrated high energy-dense structural compositions, corn (Zea mays), and switchgrass (Panicum virgatum)) and
such as glucans (26.9-31.8%), xylan (14.9-18.4%), and lignin the NUE likely improved by increasing the vegetative growth
(8.3-18.9%) [12, 16]. Sweet sorghum hybrids usually have juicy duration. Biomass sorghum’s response to N application in-
stalks with high concentrations of non-structural carbohydrates cluding the expected yield, phenotypic characteristics (e.g.,
[2, 4]. McKinley et al. [12] reported that a new PS lignocellulosic plant height, stem thickness, and leaf to stem ratio), and bio-
energy sorghum hybrid can contain up to 60% of fermentable mass composition, however, varies across growing condi-
components (dry biomass), including 50% of structural cellulose, tions, cultivar types, and cropping systems [7, 10, 31]. For
xylan, galactan, arabinan, and 10% of non-structural water-solu- new energy sorghum hybrids, it is critical to investigate the
ble sugars, that can be utilized for ethanol production via bio- effect of N practices on sorghum yield potential and feedstock
chemical conversion processes. Carbon-rich lignin, although un- composition across environmental gradients for management
desirable for biochemical conversion, accounts for another 10% optimization and sustainable feedstock production. The spe-
and can be utilized for producing biofuel such as crude bio-oil cific objectives were to (1) evaluate the yield potential of four
using thermochemical conversion techniques, such as fast pyrol- PS sorghum hybrids, and (2) determine the optimum N rate by
ysis. In contrast, ash (inorganic compounds) content is a hin- evaluating the effect of N fertilization on energy sorghum
drance to all conversion processes not only because ash has biomass yield, feedstock compositional characteristics, and
limited energy potential but likely reduces the conversion effec- N removal and use efficiency in two growing environments,
tiveness resulting from strong catalytic effects on the processing Illinois and Texas.
pipelines [15, 17, 18].
Variations in both biomass yield and composition of sorghum
hybrids are also substantially influenced by growing conditions Materials and Methods
and management practices [12, 16, 19–23]. Gill et al. [21] report-
ed that the production environment affected sorghum biomass Research Sites and Experimental Design
yield; yields ranged from 1.5 to 41.1 Mg DM ha−1 across seven
locations in the USA (i.e., Kansas, Iowa, Kentucky, Mississippi, Field studies were conducted at the University of Illinois
North Carolina, and College Station and Corpus Christi in Texas) Energy Farm located in Urbana, Illinois (IL, 40° 3′ N, 88°
Bioenerg. Res.

12′ W) and the Texas AgriLife Research Farm near College a dry stalk. Four N treatments with the application rate of 0,
Station (ARECCS), TX (30° 32′ N, 94° 26′ W) during the 56, 112, and 168 kg N ha−1 were evaluated for the biomass
growing seasons of 2018 and 2019 (Table 1). A temperate response. The experimental design was a split-plot using N
humid continental climate type was categorized for the IL treatments as the main plot and hybrids as the subplot within a
location with an average annual temperature of 11.5°C and randomized complete block with four replications. Each sub-
annual precipitation of 1030 mm for the past 30 years (1990- plot was 3 m by 12 m, consisting of 8 rows with 0.76-m row
2019 shown in Table 1). The soil type of the main plot in IL spacing.
was predominated by Drummer silty clay loam (fine-silty, The summary of field operations was described in Table 1.
mixed, superactive, mesic Typic Endoaquolls). The TX loca- In IL, the soybean stubble was disked prior to planting, and
tion was categorized as a subtropical climate type with an sorghum was planted at 180,000 seeds ha −1 using an
average annual temperature of 20.7°C and annual precipita- ALMACO four-row Kinze planter (Nevada, IA). In TX, the
tion of 1020 mm (30-year average). The primary soil type in prior crop was cotton; fields were cultivated and prepared in
TX was dominated by Ships clay loam (very-fine, mixed, the fall and sorghum was planted at 180,000 seeds ha−1 using
active, thermic Chromic Hapluderts). Table 2 shows the addi- an ALMACO four-row Max-emerge planter. Both locations
tional soil-related information in both locations and years. The had the same N fertility program. Aqueous urea ammonium
IL site typically has longer day lengths than the TX site during nitrate (UAN: 32-0-0) was knife-injected approximately a
the growing season (Fig. 1). The local monthly precipitation week after planting in IL and on the same day, post-planting
and temperature in 2018 and 2019, along with 30-year aver- in TX. The biomass harvest timing was based on site obser-
ages (1990-2019), were obtained from the National Oceanic vations, when 50% of sorghum plants displayed visible inflo-
and Atmospheric Administration for the IL site (Champaign- rescences (heading). For both locations, the plots were har-
Urbana Willard Airport, USW00094870) and the TX site vested by cutting 4 rows in the middle of each plot using a
(College Station Easterwood Field, USW00003904). John Deere (Moline, IL) forage harvester with a TCI Research
Four hybrids of PS energy sorghum were selected for this Plot Sampler 130S (Williamstown, WI).
study: TX08001, TX17500, TX17600, and TX17800. The
TX08001 and TX17500 hybrids initiate reproductive growth Plant Tissue and Chemical Composition Analysis
when day length is less than 12’20” and are categorized as
very PS (VPS) hybrids; these hybrids will not flower in IL Biomass tissue samples were directly collected from the TCI
because freezing temperatures occur before flowering would Research Plot Sampler 130S (Williamstown, WI) during har-
occur. Compared to VPS hybrids, the TX17600 and TX17800 vest, and fresh sample weight was recorded. The biomass
are less sensitive to this day length threshold and will initiate subsamples were dried in an oven for 3 days at 60°C to mea-
reproductive growth across a range of day lengths (12’40” to sure dry matter content and were saved for tissue chemical
13’00”); therefore, they are categorized as moderate PS (MPS) composition analysis. Dried tissue samples were ground
hybrids. The phenotypic characteristics of four hybrids col- through a 2-mm screen in order to homogenize the sample
lected from College Station, TX are also shown in Table 3. and two subsamples were collected. One sample was analyzed
Generally, the four PS hybrids have a good standability with for total N by a dry combustion method using a LECO FP-528
tall stalks ranging from 3 to 5 m. The four hybrids have dif- N/Protein Determinator (Leco Inc., St. Joseph, MI), while
ferent stalk thicknesses ordering from TX08001 and TX17800 phosphorus (P) and potassium (K) were analyzed using a
(thick) > TX17600 (medium) > TX17500 (thin). Both PerkinElmer Optima 8300 inductively coupled plasma (ICP)
TX17500 and TX17600 have juicy stalks, and TX17800 has spectroscopy (PerkinElmer, Inc., Waltham, MA) following a

Table 1 The experimental locations, environmental conditions, and field activities, including dates for planting and N fertilizer application for the
study period

Location 30-year average (annual) Year Planting N appl. Harvest Growth period Cumulative day GDD
(days) length (h)

Precip. (mm) Temp. (°C) VPS MPS VPS MPS VPS MPS VPS MPS

Urbana, IL 1030 11.5 2018 May 18 May 22 Oct. 09 Sep. 13 144 118 1998 1687 1754 1470
(40° 3′ N, 88° 12′ W) 2019 Jun. 01 Jun. 07 Oct. 09 Sep. 19 130 110 1793 1556 1509 1314
College Station, TX 1020 20.7 2018 May 03 May 03 Sep. 20 Sep. 20 140 140 1898 1898 2451 2451
(30° 32′ N, 94° 26′ W) 2019 May 16 May 16 Sep. 17 Sep. 17 124 124 1685 1685 2219 2219

Precip. annual precipitation, Temp. annual mean temperature, N appl. N fertilizer application with the rate of 0, 56, 112, and 168 kg N ha−1 , VPS very
photoperiod sensitive, MPS moderate photoperiod sensitive, GDD growing degree days with a base temperature of 11°C accumulated during the growth period
 Bioenerg. Res.

Table 2 Soil characteristics at the

depth of 0-30 cm (shown as mean Location Year Soil pH Organic matter CEC P K
−1 −1
± standard error) for each growing (%) (meq 100g ) (kg ha ) (kg ha−1)
environment, including two ex-
perimental sites of IL and TX Urbana, IL 2018 6.4±0.0 5.1±0.3 20.4±0.3 58.3±18.3 169.3±22.9
during the growing seasons of 2019 6.7±0.2 4.5±0.3 19.4±0.2 79.6±17.4 213.0±27.5
2018 and 2019
College Station, TX 2018 8.0±0.0 3.5±0.0 55.4±0.7 67.3±29.3 758.8
±108.9
2019 8.1±0.0 3.2±0.2 53.2±0.2 71.8±18.3 777.9±89.7
Methods - Water Loss on ignition - Mehlich 3 Mehlich 3

concentrated HNO3 and HCl microwave digestion procedure agronomic NUE and was calculated based on Eq. 2. The
using a MARSXpress vessel (CEM, Matthews, NC). Nutrient amount of N in crop biomass (or crop nitrogen removal) that
and mineral removal by sorghum biomass was calculated by was originated from the applied N was defined as the N re-
multiplying biomass yield by the corresponding concentration covery efficiency (NRE) shown in Eq. 3.
of each element (kg ha−1). The other samples were analyzed

BYtreatment
for feedstock chemical composition using the near-infrared PNUE kg kg−1 ¼ ð1Þ
spectroscopy (NIRS) compositional prediction model devel- TNUtreatment
oped by the National Renewable Energy Laboratory, and the
BY treatment −BYcontrol
NIE kg kg−1 ¼ ð2Þ
analytical process was modified based on the published pro- ΔN applied
tocol [32]. The NIRS analysis provides a more detailed insight TNUtreatment −TNUcontrol
on biomass composition, including structural glucan, xylan, NREð%Þ ¼  100% ð3Þ
ΔN applied
galactan, arabinan, acetyl, protein, and non-structural ash, wa-
ter (e.g., sucrose), and ethanol extractives (e.g., chlorophyll BYtreatment and BYcontrol (or TNUtreatment and TNUcontrol)
and waxes). are BY (or TNU) for different N treatments and control plots
(zero N input), respectively. The ΔNapplied is the difference in
Nitrogen Use Efficiency UAN-N applied between treatments and control [35].

The entire plant aboveground biomass was harvested for the Statistical Analysis
estimations of biomass yield, N uptake/removal, and N use
efficiency. Physiological N use efficiency (PNUE) was deter- Statistical analysis with PROC MIXED procedure in SAS
mined by a ratio of DM biomass yield (BY) to the total N [36] was used to test the effect of locations (IL and TX),
uptake (TNU), shown on Eq. 1 [31, 33, 34]. The increased harvest year (2018 and 2019), N rates (four rates), hybrids
biomass yield per applied unit of N fertilizer was defined as (four hybrids) and their interactions on biomass yield produc-
the N input efficiency (NIE), which is often referred to the tion, plant tissue nutrient concentrations and removal, NUE,
and feedstock chemical compositions using analysis of vari-
ance (ANOVA). Locations, hybrid, N rates, harvest year, and
their interactions were considered fixed effects, where repli-
cations and their interactions were considered random effects.
All significant differences were determined at P ≤ 0.05.
Pairwise mean comparisons were made using the Fisher’s
least significant difference (LSD) test.

Results

Environment

Monthly precipitation and temperature during the study period

(2018 and 2019) and their 30-year average (1990-2019) for
two experimental sites (Urbana, IL, and College Station, TX)
Fig. 1 Day light hours at Urbana (IL) and College Station (TX) in 2019 are shown in Fig. 2. During the growing season from May
Bioenerg. Res.

Table 3 Phenotypic information

of four photoperiod-sensitive en- Maturity Hybrid Standability Height (m) Stalk thickness (mm) Stalk composition
ergy sorghum hybrids
VPS TX08001 Good Tall Thick Intermediate
TX17500 Excellent Tall Thin Juicy
MPS TX17600 Good Tall Medium Juicy
TX17800 Good Tall Thick Dry

VPS very photoperiod sensitive, MPS moderate photoperiod sensitive

(planting) to Oct (harvest), the cumulative precipitation was precipitation only contributed approximately 27% to the
higher in 2018 than 2019 and the 30-year average in both May-Oct cumulative precipitation (142 mm out of 527 mm)
locations (IL: 676 mm in 2018 and 504 mm in 2019; TX: in 2019, showing a water stress condition in the late growing
657 mm in 2018 and 527 mm in 2019). The monthly temper- season. This variation is typical in Texas and is associated
ature was similar between 2018 and 2019 for both IL and TX. with tropical storms from Aug to Oct. In IL, the sum of pre-
Compared to the 30-year averages, the precipitation pattern cipitation in Sept and Oct accounted for 24% and 28% of the
showed more fluctuations than the temperature. In TX, the overall precipitation during the growing season in 2018 and
cumulative precipitation during the growing season (May- 2019, respectively, both of which showed a similar trend to
Oct) was predominated by the precipitation in Sept (210 the 30-average results (a total of 164 mm in Sept and Oct out
mm) and Oct (298 mm) in 2018, accounting for 77% of 600 mm from May to Oct). The soil chemical properties
(508 mm out of 657 mm); by contrast, the Sept and Oct were also different in both locations. The TX site showed

Fig. 2 Local weather conditions at the experimental sites across the 2 years of the study including a monthly precipitation and b average monthly
temperature and the 30-year monthly average (1990-2019) (data: NOAA)
 Bioenerg. Res.

higher soil pH (8.0-8.1), CEC (53.2-55.4 meq 100g−1), and K 15%, respectively, compared to the yield from TX. The aver-
content (758.8-777.9 kg ha−1) but lower organic matter (3.2- age across the 2 years and four hybrids showed that PS sor-
3.5%) than the IL site having soil pH ranging from 6.4 to 6.7, ghum had higher yield potential in IL (20.4 Mg DM ha−1) than
CEC from 19.4 to 20.4 meq 100g−1, soil K from 169.3 to in TX (19.0 Mg DM ha−1) (Table 5). The interaction between
213.0 kg ha−1, and organic matter from 4.5 to 5.1% (Table 2). N rate and other factors (location, year, and hybrid) did not
show any significant impacts on biomass yield (Table 4).
Biomass Yield Increased N rate generally increased the sorghum biomass
yield (Table 4). However, the biomass yields of the four hy-
The effects of four factors (location, year, N rate, and hybrid) brids showed similar responses to the increased N rate, with
on biomass yield are shown in the ANOVA results (Table 4). no benefits observed from N rates above 112 kg ha −1
The four-way interaction among these factors was not signif- (Table 6).
icant for biomass yield; however, the three-way interaction of
location, year, and hybrid significantly influenced sorghum Nutrient Concentrations and Removal
biomass yield (Table 4). The three-way interaction was caused
by low biomass yield and variation among hybrids in TX- The three-way interaction between location, year, and hybrid
2018 (Fig. 3a). Except for TX17600, the yields of TX08001, was also significant for tissue N, P, and K concentrations
TX17500, and TX17800 were significantly lower in TX-2018 (Table 4). In each environment, all hybrids showed similar
compared to the other three environments (IL-2018, IL-2019, tissue N concentrations except for biomass in TX-2018, which
and TX-2019). Contrasted to yield averaged across IL-2018, had higher variations in N concentrations (Fig. 3b). Averaged
IL-2019, and TX-2019, the biomass yield in TX-2018 was across the four hybrids, N concentrations were generally
lower by 21%, 47%, and 21% for TX08001, TX17500, and higher in 2019 (8.7 g kg−1) than in 2018 (7.0 g kg−1) in IL;
TX17800, respectively. The two-way interaction between lo- by contrast, increased N concentrations were observed in
cation and hybrid also affected biomass yield (Table 4), and 2018 (8.7 g kg−1) compared to the concentrations in 2019
the 2-year average showed that TX08001, TX17500, and (7.8 g kg−1) in TX. This was likely due to the high N concen-
TX17800 hybrids had higher yield potential in IL than in trations of the TX17500 (10.4 g kg−1) and TX17800 (9.8 g
TX except for TX176000 (Table 5). In IL, the TX08001, kg−1) hybrids in 2018 (Fig. 3b). For P concentrations, the
TX17500, and TX17800 biomass yield averaged across TX17500 hybrid tended to have higher P concentrations than
2018 and 2019 were significantly higher by 10%, 22%, and other hybrids in both locations, although differences were

Table 4 Analysis of variance (ANOVA) showed the effects of main removal, and indices of physiological N use efficiency (PNUE), N input
factors of location (L), year (Y), N rate (Nr), and hybrids (H) and their efficiency (NIE), and N recovery efficiency (NRE)
interactions on biomass yield, biomass nutrient concentration, nutrient

Effect Yield Plant nutrient concentration Plant nutrient removal Efficiency index

N P K N P K PNUE NIE NRE

L 0.0018* 0.1466 <.0001* <.0001* 0.0591 <.0001* <.0001* 0.1081 0.5924 0.6553
Y <.0001* 0.0461* <.0001* 0.7168 <.0001* 0.3914 <.0001* 0.0157* 0.0015* 0.0050*
Nr 0.0009* 0.0098* 0.1846 0.9016 0.0015* 0.0807 0.0936 0.0133* 0.1111 0.7951
H 0.0003* 0.0064* <.0001* 0.0016* 0.1081 0.4298 0.1117 0.0098* 0.1019 0.6568
L×Y <.0001* <.0001* 0.0090* <.0001* 0.1246 <.0001* <.0001* <.0001* 0.0020* 0.5527
L × Nr 0.7042 0.0148* 0.2967 0.0061* 0.0089* 0.0839 0.0394* 0.0759 0.9336 0.1369
L×H <.0001* 0.0036* 0.1309 0.0406* 0.1987 0.0123* <.0001* 0.0096* 0.0604 0.1642
Y × Nr 0.1421 0.9391 0.1085 0.2448 0.1934 0.0821 0.0009* 0.9708 0.4437 0.6996
Y×H 0.0001* 0.020* 0.0040* 0.0006* 0.0142* 0.5053 0.0159* 0.0350* 0.4846 0.3117
H × Nr 0.2865 0.9829 0.5137 0.0253* 0.9361 0.895 0.1392 0.9010 0.9452 0.9809
L × Y × Nr 0.1823 0.0940 0.581 0.1947 0.8557 0.8916 0.073 0.1717 0.4873 0.8456
L×Y×H 0.0020* 0.0016* 0.0362* 0.0009* 0.2818 0.1659 0.0140* 0.0145* 0.0752 0.4543
L × Nr × H 0.6587 0.9374 0.7412 0.2127 0.5951 0.6556 0.5299 0.9295 0.9637 0.8999
Y × Nr × H 0.8655 0.8764 0.2792 0.0635 0.6477 0.8527 0.0676 0.7576 0.8567 0.9710
L × Y × Nr × H 0.3477 0.9025 0.0893 0.291 0.8725 0.4621 0.0986 0.9223 0.3761 0.9115

*Significant effect at the level of P < 0.05

Bioenerg. Res.

Table 5 The significant effect of location (IL and TX) × hybrid standard error in parentheses. Lowercase letters indicate mean separation
(TX08001, TX17500, TX17600, and TX17800) on dry biomass yields, α=0.05 organized highest to lowest value for each column (no mean
nutrient concentrations, and removal (N, P, and K), shown as mean and separations were applied if the variable effect was not significant)

Location IL TX Location mean

Hybrid TX08001 TX17500 TX17600 TX17800 TX08001 TX17500 TX17600 TX17800 IL TX

Dry biomass (Mg ha−1)

Yield 20.9ab 20.2bc 19.5bcd 20.9ab 18.9cd 16.5e 22.5a 18.3d 20.4a 19.0b
(0.7) (0.7) (0.6) (0.9) (0.7) (1.2) (0.6) (0.6) (0.4) (0.4)
Biomass nutrient concentration (g kg−1)
N 7.3d 7.7bcd 8.4abc 8.3abc 7.7bcd 9.2a 7.6cd 8.5ab 7.9 8.3
(0.3) (0.4) (0.4) (0.4) (0.4) (0.4) (0.4) (0.3) (0.2) (0.2)
P 1.4 1.6 1.4 1.3 1.3 1.4 1.2 1.3 1.4a 1.3b
(0.0) (0.1) (0.0) (0.0) (0.1) (0.1) (0.1) (0.1) (0.0) (0.0)
K 11.7d 13.2c 10.5de 10.2e 20.4ab 20.3ab 18.9b 20.6a 11.4b 20.1a
(0.4) (0.5) (0.3) (0.3) (0.7) (0.8) (1.0) (0.9) (0.2) (0.4)
Plant nutrient removal (kg ha−1)
N 149.9 158.4 159.9 175.6 142.7 136.3 167 147.3 160.9 148.3
(8.0) (11.4) (9.5) (11.2) (9.7) (11.8) (10.9) (6.8) (5.1) (5.0)
P 29.4ab 31.6a 27.8bc 27.7bc 24.2cde 21.9e 26.3bcd 23de 29.1a 23.9b
(1.3) (1.5) (1.1) (1.3) (1.5) (1.8) (1.5) (0.9) (0.7) (0.7)
K 243.7de 261.9d 203.5f 213.5ef 384.4b 323.9c 427.8a 367.6b 230.6b 375.8a
(10.8) (10.1) (7.9) (9.9) (18.9) (32.2) (29.1) (24.1) (5.2) (13.5)

significant only in 2018. Across all environments, P concen- than the average of the other three sorghum hybrids. All hy-
trations in TX17500 biomass were approximately 13% higher brids had higher tissue K concentrations in TX than in IL. The

Table 6 The significant 2-way interaction between location (IL and in parentheses. Lowercase letters indicate mean separation α=0.05 orga-
TX) and N rate (0, 56, 112, and 168 kg N ha−1) on biomass yields, nized highest to lowest value (no mean separations were applied if the
nutrient concentrations, and removal, shown as mean and standard error location × N rate interaction was not significant)

Location IL TX

N rate 0 56 112 168 0 56 112 168

Dry biomass (Mg ha−1)

Yield† 18.1 19.8 21.5 22.1 17.1 18.9 20.1 20.0
(0.8) (0.5) (0.8) (0.5) (0.9) (0.9) (0.8) (0.9)
Biomass nutrient concentration (g kg−1)
N 6.9d 7.2d 7.8bcd 9.5a 7.5cd 8.5abc 8.3bc 8.7ab
(0.3) (0.3) (0.4) (0.3) (0.4) (0.5) (0.4) (0.3)
†
P 1.4 1.5 1.4 1.4 1.4 1.4 1.2 1.2
(0.1) (0.1) (0.0) (0.0) (0.1) (0.1) (0.1) (0.0)
K 12.1bc 12.3b 10.7bc 10.6c 19.5a 19.5a 20.2a 21.0a
(0.5) (0.4) (0.3) (0.4) (0.8) (0.9) (0.8) (0.9)
Plant nutrient removal (kg ha−1)
N 126.7d 141.7cd 166.2b 209.2a 125.1d 149.2bcd 159.6bc 159.3bc
(8.1) (7.1) (10.9) (7.1) (9.1) (11.0) (9.8) (8.5)
P† 25.8 29.4 29.7 31.6 23.2 24.8 24.8 22.6
(1.5) (1.3) (1.3) (1.0) (1.5) (1.4) (1.6) (1.4)
K 217.9c 242.5c 227.6c 234.4c(8.7) 330.4b 365.5ab 402.9a 404.3a
(11.5) (10.9) (10.5) (22.3) (30.0) (25.5) (27.2)
†
No significant location × N rate interaction effect
 Bioenerg. Res.

a TX08001 TX17500 TX17600 TX17800 averages across all hybrids and the 2 years showed that the PS
30 sorghum grown in TX had increased tissue K concentrations
LSD0.05 = 2.4
by approximately 75% (Table 5). For nutrient removal, the
25 interaction among location, year, and hybrid only affected
the biomass K removal, but the location × hybrid interaction
Yield (Mg DM ha-1)

20 influenced both P and K removal (Table 4). No consistent

trend was observed for P and K removal among hybrids in
both locations, but the average across hybrids and years
15
showed that higher P removal occurred in IL (29.1 kg ha−1)

23.6
than in TX (23.9 kg ha−1). Conversely, the PS sorghum had a
22.7
22.3

21.9
21.4

21.4
20.3

20.2
20.1
19.6
19.5
19.3

19.2
10
63% higher K removal in TX compared to the sorghum grown
16.5

16.3
11.0
in IL (Table 5).
5 Although the four hybrids had similar yield responses to
the N rate (no interaction between N rate and other variables),
0 the interaction of location and N rate was significant for the
2018 2019 2018 2019
plant tissue N and K concentrations as well as their corre-
IL TX
sponding plant nutrient removal (Table 4). Both N and K
concentrations responded differently to N rate in IL and TX.
b TX08001 TX17500 TX17600 TX17800
The tissue N concentration increased with increasing N rate in
12.5 LSD0.05 = 1.2 both locations (Table 6). Compared to the control zero N
treatment, the highest N input of 168 kg ha−1 increased the
10.0 N concentrations by approximately 38% in IL but only by
16% in TX (Table 6). In contrast, the 168 kg ha−1 N rate
N (g kg-1)

7.5
reduced the tissue K concentration by 14% compared to the
56 kg ha−1 N input, which was only observed in IL. The tissue
10.4

9.8

5.0
9.2
8.9
8.8

8.2
8.0
8.0

7.7
7.6

7.4

P concentrations did not respond to N fertilizer rate. For nu-

7.3
7.2
7.2
6.7
6.6

2.5 trient removal, the ANOVA results also showed that the loca-
tion × N rate interaction influenced the crop N and K removal
0.0
LSD0.05 = 0.2
(Table 4). The response of the biomass N removal to the
2.0 increased N input was more substantial in IL than in TX.
Compared to the zero N input, the 168 kg ha−1 N rate resulted
1.5 in a 65% increase in N removal in IL but only 27% in TX
P (g kg-1)

(Table 6). Based on the linear relationship between N rate and

1.0 removal, approximately an additional 0.5 kg ha−1 and 0.2 kg
1.8

1.6
1.6

ha−1 of N were removed from the IL (R2 = 0.95) and TX (R2 =

1.5

1.4

1.4

1.3

1.3
1.3

1.3

1.3
1.3

1.2
1.2

1.2
1.1

0.5 0.81) soils, respectively, by increasing 1.0 kg N ha−1 input. A

significant response of K removal to N rate was only observed
0.0 in TX. With an increasing N rate from 0 to 168 kg ha−1, the
LSD0.05 = 2.0
25 biomass K removal increased by nearly 22% with the remov-
ing rate of 0.5 kg K ha−1 per 1.0 kg N ha−1 input (R2 = 0.90).
20
No significant location × N rate effect on biomass P concen-
K (g kg-1)

15
trations or removal was observed in this study.
22.6
20.9
20.8
20.6

20.3
20.1

20.1

10 Nitrogen Use Efficiency

15.2
14.7
12.4

11.6
11.4

11.1
10.7

10.0
9.5

5
Three NUE indices (PNUE, NIE, and NRE) used in this study
0 were also influenced by the environment. The location × year
2018 2019 2018 2019
interaction was significant for PNUE and NIE. The highest
PNUE appeared in the IL-2018 (151.3 kg kg−1), among
IL TX

Fig. 3 The significant effect of the 3-way interaction among location (IL others, but there was no year effect on PNUE in TX
and TX), year (2018 and 2019), and hybrid (TX08001, TX17500,
TX17600, and TX17800) on a annual dry biomass (DM) yields and b
(Table 7). For the NIE, the IL location showed consistent
biomass tissue nutrient (N, P, K) concentrations of energy sorghum. Error NIE in both years (29.5 kg DM improvement per 1.0 kg N
bars represent the standard error of the means input); however, approximately a 19-fold difference between
Bioenerg. Res.

2018 and 2019 was observed in TX (2.6 kg DM in 2018 vs. showed similar compositions within a growing environment,
48.8 kg DM per 1.0 kg N input shown in Table 7). The aver- but the concentrations differed across environments (Fig. 4).
age across the two locations indicated that the NIE was higher Across the two experimental years, the effects of the 2-way
in 2019 than in 2018, so was NRE (18.1% in 2018 vs. 51.1% interaction between locations and hybrids on the biomass
in 2019). The N effect on the three NUE indices, however, structural and soluble components are shown in Table 9. For
only influenced the PNUE. Although the location × N rate both IL and TX, the TX17800 hybrid consistently contained
interaction showed a weak impact on PNUE at the P < 0.05 higher concentrations of the structural glucan, xylan, and lig-
threshold (P = 0.0759), the PNUE reduced with increasing N nin than other hybrids; likewise, TX17800 tended to have
input in both IL and TX (Table 7). Compared to the zero N higher concentrations for both biochemical-processing and
treatment, the highest N rate (168 kg ha−1) resulted in declines thermochemical-processing interested components (i.e., BIC
in PNUE by around 25% and 17% in IL and TX, respectively. and TIC). For instance, the BIC and TIC concentrations of the
Neither the location × N rate interaction nor N rate showed a TX17800 hybrid were approximately 5% and 7% higher, re-
significant impact on NIE and NRE, but the overall trend spectively, compared to other hybrids (Table 9). In contrast,
showed that the increased N input likely reduced both NIE TX17800 had the lowest concentrations of soluble sucrose
and NRE. and other non-structural inorganics (e.g., ash). TX17600 had
the highest sucrose concentrations among hybrids by approx-
Chemical Composition imately 34% and 89% in IL and TX, respectively (Table 9).
For other structural components, TX08001 (VPS hybrid) had
Sorghum feedstock chemical composition was influenced by the highest concentrations of galactan, and TX17800 (MPS
the environment, hybrid, N rate, and their interactions hybrid) had the lowest arabinan among hybrids. The VPS
(Table 8). The 4-way interaction of location, year, N rate, hybrid TX17500 had the highest biomass ash concentration
and hybrid was not significant for the biomass structural and among hybrids. Between the two locations, the TX site gen-
soluble compositions; however, the 3-way interaction among erally had higher concentrations of BIC, TIC, and acetyl, cor-
location, year, and hybrid showed a significant impact on the responding to the lower ash concentrations than observed at
concentrations of structural and soluble components (Table 8). the IL site (Table 9). The N effect on feedstock composition
All hybrids in TX-2019 had the lowest structural component showed that only a few chemical components, including lig-
(a sum of glucan, xylan, lignin, galactan, arabinan, acetyl, and nin, acetyl, ash, and sucrose, responded significantly to the N
protein) concentrations among the three environments (IL- rate for all hybrids and no interaction of N rate × hybrid was
2018, IL-2019, and TX-2018) and TX17600 tended to have observed (Table 8). The increased N rate increased lignin and
the lowest structural component concentration among all hy- ash concentrations but reduced acetyl and sucrose concentra-
brids (Fig. 4). The two VPS hybrids (TX08001 and TX17500) tions (Table 10).

Table 7 The significant 2-way

interaction between (a) location (a) Location × year
(IL and TX) and year (2018 and Year PNUE (kg kg−1) NIE (kg kg−1) NRE (%)
2019) and (b) location and N rate IL TX Mean IL TX Mean IL TX Mean
(0, 56, 112, and 168 kg N ha−1) on
indices of physiological N use 2018 151.3a 122.1bc 136.7A 29.2a 2.6b 15.9B 22.9 13.4 18.1B
efficiency (PNUE), N input effi- (5.0) (6.0) (4.0) (8.2) (4.7) (4.8) (8.8) (7.1) (6.2)
ciency (NIE), and N recovery ef- 2019 118.6c 134.1b 126.3B 29.9a 48.8a 39.3A 50.5 51.7 51.1A
ficiency (NRE), shown as mean
(3.3) (3.9) (2.6) (6.4) (8.4) (5.4) (9.0) (11.0) (7.0)
and standard error in parentheses.
Lowercase and uppercase letters (b) Location × N rate
indicate mean separation α=0.05 N rate PNUE† (kg kg−1) NIE (kg kg−1) NRE (%)
organized highest to lowest value IL TX Mean IL TX Mean IL TX Mean
for year × location and year, re-
0 147.8a 142.5ab 145.1A - - - - - -
spectively (no mean separations
were applied if the variable effect (5.5) (7.7) (4.5)
was not significant) 56 148.0a 126.1bcd 137.1A 33.3 33.1 33.0 25.9 47.5 38.3
(6.9) (7.6) (5.2) (11.3) (12.9) (8.5) (14.5) (18.9) (11.6)
112 133.7abc 125.0bcd 129.4AB 31.8 27.1 29.1 35.7 32.9 33.1
(7.0) (5.5) (4.6) (9.3) (7.8) (6.0) (11.3) (11.3) (8.0)
168 110.2d 118.7cd 114.4B 23.6 16.9 20.5 48.5 17.3 32.9
(4.4) (4.2) (3.2) (5.3) (5.4) (3.8) (6.4) (7.9) (5.2)
†
Weak location × N rate interaction effect (P = 0.0759)
 Bioenerg. Res.

Table 8 Analysis of variance (ANOVA) showed the effects of main factors of location (L), year (Y), N rate (Nr), and hybrids (H) and their interactions
on feedstock chemical composition

Effect Overall Structural component Non-structural

Structural Soluble Glucan Xylan Lignin Galactan Arabinan Acetyl BIC TIC Ash Sucrose

L <.0001* <.0001* <.0001* <.0001* 0.3462 <.0001* <.0001* <.0001* <.0001* 0.0006* <.0001* <.0001*
Y <.0001* <.0001* <.0001* <.0001* <.0001* 0.0232* 0.7429 <.0001* <.0001* <.0001* <.0001* <.0001*
Nr 0.0039* 0.0052* 0.6754 0.1660 0.0002* 0.0556 0.0628 0.0004* 0.6977 0.0859 0.0203* 0.0103*
H <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.2014 <.0001* <.0001* <.0001* <.0001*
L×Y <.0001* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0286* <.0001* <.0001* <.0001* <.0001* <.0001*
L × Nr 0.6628 0.4315 0.1560 0.1254 0.3758 0.0009* 0.0888 0.0005* 0.1723 0.1877 0.0055* 0.3442
L×H <.0001* <.0001* 0.0012* <.0001* <.0001* 0.0005* 0.0001* 0.8495 <.0001* <.0001* 0.0535 <.0001*
Y × Nr 0.2326 0.1873 0.0063* 0.6081 0.1117 0.1731 0.1566 0.0206* 0.0653 0.0482* 0.0027* 0.0755
Y×H 0.1153 0.2670 0.2402 0.0729 0.0022* 0.0016* 0.1329 0.0003* 0.4110 0.3354 0.0958 0.0148*
H × Nr 0.5899 0.6126 0.7186 0.7475 0.7721 0.1663 0.9291 0.0439* 0.7513 0.7427 0.2233 0.6013
L × Y × Nr 0.3184 0.2875 0.4212 0.5163 0.2578 0.0093* 0.3802 <.0001* 0.5393 0.6150 <.0001* 0.0279*
L×Y×H 0.0034 0.0103* 0.0111* <.0001* <.0001* <.0001* <.0001* <.0001* 0.0085* 0.0006* <.0001* 0.0654
L × Nr × H 0.1792 0.1645 0.3774 0.2708 0.5008 0.0525 0.6378 0.1665 0.2993 0.3005 0.1716 0.4745
Y × Nr × H 0.2834 0.2930 0.3393 0.0779 0.0210* 0.2972 0.7668 0.8974 0.1586 0.0650 0.9801 0.1985
L × Y × Nr × H 0.4240 0.4105 0.3025 0.4866 0.3458 0.8952 0.7484 0.6408 0.3647 0.3081 0.4405 0.5049

BIC biochemical-processing interested components based on the sum of the structural glucan, xylan, galactan, and arabinan, TIC thermochemical-
processing interested components based on the sum of the BIC and structural lignin
*Significant effect at the level of P < 0.05

Discussion yield potential (2-year average) in TX (22.5 Mg DM ha−1)

than in IL (19.5 Mg DM ha−1), which was possibly due to
Environment Effects the prolonged growth period in TX (Table 1); by contrast, the
TX17800 hybrid showed higher yield potential in IL.
The cultivation environment plays a significant role in bio- Weather variations also influenced biomass yield signifi-
mass yield, feedstock quality, and the strategies for optimizing cantly [10]. The monthly temperature in IL and TX was con-
management practices [3, 11, 20, 21, 23, 37]. Four environ- sistent in both the experiment years compared to the 30-year
mental regimes (two locations by 2 years) were evaluated in averages; however, the precipitation was highly variable (Fig.
this study. Results showed highly significant interactions be- 2). The great variability in precipitation not only increases the
tween the environment and other factors for the yield poten- difficulty in optimizing field management but also influences
tial, biomass nutrient concentrations and removal, and feed- crop development and biomass production [39–41]. In TX,
stock compositions, indicating different responses to hybrids the seasonal precipitation was concentrated during the late
and N treatments among environments. These environmental growing season in 2018, whereas during the early season in
variations included differences in daylight length, soil proper- 2019 (Fig. 2), the highly fluctuating precipitation pattern like-
ties, and most importantly, weather. ly led to more yield variation in TX than in IL (Fig. 3). For
The PS nature of these hybrids keeps them in the vegetative instance, both VPS hybrids and TX17800 had lower biomass
stage and allows them to continuously accumulate biomass yields in the TX-2018 regimes than in the other three environ-
yield by increasing both leaf area index and solar radiation ments. This yield decline was attributed to the drought condi-
inception during the growing season [5, 7, 11, 38]. tion during the early growing season from May to Aug (2018:
Compared to the TX site, the increased biomass yield of two 150 mm vs. 30-year average: 324 mm) in TX of 2018.
VPS hybrids grown at the IL site (TX08001 and TX17500 Inadequate rainfall slows the crop’s early establishment and
shown in Table 5) was attributed to several potential factors delays growth and yield [21, 40]. In this study, the overall
including longer days at the peak of the growing season, cool- biomass yield of the PS sorghum was similar to our previous
er night temperatures, and more consistent rainfall in the results reported by Maughan et al. [10], but approximately
growing season (Table 1). The two MPS hybrids (TX17600 30% lower than the other reports (>30 Mg DM ha−1) where
and TX17800) responded differently to the growth period/ the field trials were mostly in the long-growing season (> 150
cumulative day length. The TX17600 hybrid exhibited higher days) areas [11–13]. The relatively low yield was presumably
Bioenerg. Res.

Table 9 The significant effects of location (IL and TX) and hybrid parentheses. Lowercase letters indicate mean separation α=0.05
(TX08001, TX17500, TX17600, and TX17800) on concentrations of organized highest to lowest value (no mean separations were applied if
feedstock components, shown as mean and standard error in the variable effect was not significant)

Location Hybrid Structural components (g kg−1) Non-structural (g kg−1)

Glucan Xylan Lignin Gal Arab Acetyl BIC TIC Ash† Sucrose

IL TX08001 285.4ef 166.2a 115.7c 11.5b 25.9b 12.5 489.0d 604.7d 94.3b 57.7de
(3.0) (1.6) (3.3) (0.1) (0.2) (0.3) (4.3) (7.4) (0.8) (4.4)
TX17500 279.3f 162.7b 108.0d 11.1c 26.0b 12.1 479.1e 587.1e 99.6a 62.3cd
(3.1) (1.5) (3.2) (0.2) (0.2) (0.3) (4.2) (7.2) (1.1) (4.2)
TX17600 292.0de 153.8d 115.6c 10.7d 24.2c 12.6 480.5e 596.2de 90.6b 73.1b
(3.4) (1.4) (2.3) (0.1) (0.4) (0.4) (4.0) (6.1) (1.1) (4.5)
TX17800 305.5b 168.2a 126.8a 10.7d 22.7d 12.3 507.1b 633.9b 93.7b 43.8f
(3.1) (1.3) (2.1) (0.1) (0.3) (0.3) (4.1) (5.9) (1.2) (4.2)
TX TX08001 297.2cd 159.0c 120.0b 11.9a 27.7a 16.8 495.9cd 615.9c 85.6d 65.9c
(2.6) (0.9) (1.4) (0.2) (0.3) (0.3) (2.8) (3.6) (2.2) (5.5)
TX17500 299.1bc 163.1b 118.7bc 11.5b 27.0a 16.5 500.7bc 619.4c 86.8c 56.6de
(3.1) (0.8) (1.2) (0.2) (0.4) (0.5) (3.2) (3.7) (3.1) (6.9)
TX17600 292.3d 142.4e 102.4e 11.6b 27.6a 16.9 473.9e 576.2f 81.3e 110.3a
(2.9) (1.4) (2.0) (0.2) (0.4) (0.4) (3.5) (5.1) (2.1) (6.1)
TX17800 315.5a 166.1a 128.4a 11.5b 26.0b 17 519.3a 647.5a 82.9d 51.9e
(2.7) (1.0) (1.6) (0.1) (0.4) (0.4) (3.1) (4.4) (2.6) (7.1)
Hybrid mean TX08001 291.3b 162.6b 117.9b 11.7a 26.8a 14.6 492.4b 610.3b 90.0b 61.8b
(2.1) (1.0) (1.8) (0.1) (0.2) (0.4) (2.6) (2.6) (1.4) (3.6)
TX17500 289.2b 162.9b 113.3c 11.3b 26.5ab 14.3 489.9b 603.2b 93.2a 59.4b
(2.5) (0.9) (1.8) (0.1) (0.2) (0.4) (3.0) (4.5) (1.8) (4.0)
TX17600 292.1b 148.1c 109.0d 11.2bc 25.9b 14.8 477.2c 586.2c 86.0c 91.7a
(2.2) (1.2) (1.8) (0.1) (0.4) (0.4) (2.7) (4.1) (1.6) (4.5)
TX17800 310.5a 167.2a 127.6a 11.1c 24.4c 14.6 513.2a 640.7a 88.3b 47.9c
(2.2) (0.8) (1.3) (0.1) (0.3) (0.4) (2.7) (3.8) (1.6) (4.1)
Location mean IL 290.5b 162.7a 116.5a 11.0b 24.7b 12.4b 488.9b 605.4b 96.0a 59.2b
(1.8) (0.9) (1.5) (0.1) (0.2) (0.2) (2.3) (3.7) (0.5) (2.4)
TX 301.0a 157.6b 117.3a 11.6a 27.1a 16.8a 497.5a 614.8a 82.6b 71.2a
(1.6) (1.0) (1.1) (0.1) (0.2) (0.2) (2.1) (3.1) (1.3) (3.8)

Gal galactan, Arab arabinan, BIC biochemical-processing interested components based on the sum of the structural glucan, xylan, galactan, and arabinan,
TIC thermochemical-processing interested components based on the sum of the BIC and structural lignin
†
Weak location × hybrid interaction effect (P = 0.0535)

Table 10 The significant effects

of N rate (0, 56, 112, and 168 kg N rate Structural components (g kg−1) Non-structural (g kg−1)
N ha−1) on concentrations of
feedstock components, shown as Glucan Xylan Lignin Gal Arab Acetyl BIC TIC Ash Sucrose
mean and standard error in
parentheses. Lowercase letters 0 293.9 159.7 114.5b 11.5 26.6 15.2a 491.6 606.2 88.1b 65.9ab
indicate mean separation α=0.05 (2.2) (1.4) (1.7) (0.1) (0.3) (0.4) (2.9) (4.3) (1.7) (4.2)
organized highest to lowest value
56 295.8 159.5 114.1b 11.2 25.9 14.9a 492.5 606.6 88.4b 68.9a
(no mean separations were
applied if the variable effect was (2.4) (1.2) (1.7) (0.1) (0.3) (0.4) (3.0) (4.5) (1.8) (4.4)
not significant) 112 297.2 160.6 118.6a 11.2 25.4 14.3b 494.5 613.0 89.5ab 64.7bc
(2.4) (1.5) (2.0) (0.1) (0.3) (0.4) (3.3) (5.2) (1.6) (5.0)
168 296.1 161.0 120.4a 11.3 25.7 14.0b 494.2 614.6 91.3a 61.2c
(2.9) (1.3) (2.0) (0.1) (0.3) (0.4) (3.6) (5.4) (1.5) (4.4)

Gal galactan, Arab arabinan, BIC biochemical-processing interested components based on the sum of the struc-
tural glucan, xylan, galactan, and arabinan, TIC thermochemical-processing interested components based on the
sum of the BIC and structural lignin
 Bioenerg. Res.

Fig. 4 The significant 3-way in- IL TX

teraction between location (IL and
TX), year (2018 and 2019), and 2018 2019 2018 2019
100
hybrid (TX08001, TX17500,

17.2

17.5
17.8
18.0

20.0
20.5
21.2

21.8
22.9

23.1
24.4

24.4

26.8
28.3
29.4
TX17600, and TX17800) and

35.2
their influence on dry biomass 80
structural and soluble component

Proportion (%)
proportions of energy sorghum
60
(*both structural and soluble Soluble
concentrations had the same Structural

82.8

82.5
82.2
82.0

80.0
79.5
78.8
Fisher’s least significant differ-

78.2
77.1

76.9
40

75.6

75.6

73.2
71.7
(LSD0.05* = 1.8)

70.6
ence (LSD) at α=0.05 because the

64.8
measured concentrations were
normalized to percent) 20

0 TX08001

TX17500

TX17600

TX17800

TX08001

TX17500

TX17600

TX17800

TX08001

TX17500

TX17600

TX17800

TX08001

TX17500

TX17600

TX17800
due to the short production seasons, which can be improved July as is typical compared to the 30-year benchmark (Fig.
by prolonging the growing season, such as planting earlier [7]. 2). Since the TX-2019 site received adequate rain early to
A substantial environmental impact on tissue nutrient con- facilitate crop establishment, the following dry season (Jul to
centrations shown in this study was also observed in several Oct) did not affect the biomass yield of the well-established
studies [29, 34, 42–45]. The decline in N concentrations was sorghum due to a good water stress resistance; however, the
usually corresponding to the increase in biomass yield, which non-structural compounds did respond to the dry environment
was attributed to the effect of a growth dilution [42, 45, 46]. [50].
This dilution effect on P and K concentrations, however, was
not observed in this study, which was also reported by Maw Hybrid Comparisons
et al. [45]. Compared to IL, the higher soil K content in TX
(approximately a 4-fold difference: Table 2) likely resulted in Among the four sorghum hybrids, no consistent yield trends
an increase in biomass K concentrations for all hybrids were observed under suitable conditions in IL-2018, IL-2019,
(Table 5). The increased concentrations of the alkali metals, and TX-2019 regimes. A strong interaction between the envi-
such as K and Na, in biomass can substantially affect energy ronment and hybrid showed that the four PS hybrids had dif-
conversion efficiency from the feedstock [15, 45, 47]. The PS ferent resistances to adverse growing conditions, such as in
sorghum grown in TX also led to an approximate 34% in- TX-2018. For instance, the TX17600 hybrid consistently pro-
crease in biomass Na concentrations compared to the sorghum duced higher biomass yields than the other hybrids in TX and
from IL (data not shown). The increased K concentration also showed less sensitivity to the environmental variation for bio-
resulted in higher K removal in TX than IL. Since nutrient mass production across the four cultivation regimes (Fig. 3).
removal is the product of biomass yield and tissue nutrient For the nutrient concentrations, the four PS hybrids showed
concentrations, the reduced yield (averaged across four hy- similar aboveground biomass N concentrations under similar
brids) led to lower K removal in 2018 than 2019 in TX and cultivation environments except for TX-2018 which had de-
the lowest P removal among for environmental regions. layed crop establishment (Fig. 3b). Excluding the data from
Feedstock compositions were also influenced by the grow- the TX-2018 environment, the range of the biomass N con-
ing environment [12, 16, 20]. For lignocellulosic bioenergy, it centrations (6.6 and 9.2 g kg−1) and the corresponding yields
is crucial to minimize the content of the non-structural com- (19.2-23.6 Mg DM ha−1) were similar to the previous studies
ponents; however, increased water stress could increase the on the fiber, sweet, and biomass sorghums under a similar N
concentrations of non-structural components for different input [34, 42, 44]. Other PS sorghum studies, conversely,
bioenergy feedstocks, such as corn stover, miscanthus, reported lower N concentrations (4.4 ~ 6.3 g kg−1) associated
switchgrass, and mixed perennial grasses [48–50]. McKinley with higher biomass yields (> 25 Mg DM ha−1) [31, 43]. The
et al. [12] reported that the soluble compounds in energy sor- improved yield, usually resulting from an extended growth
ghum stem biomass were generally higher, along with in- period (e.g., > 140 growth days), diluted the biomass N con-
creased sucrose, in rainfed systems than in irrigated systems. centrations [31, 42, 43]. Although the four PS hybrids had
For these four hybrids, significantly increased concentrations similar N requirements, the TX17500 had a higher stover P
of the soluble components were observed in the TX-2019 concentration, and was therefore associated with more remov-
regimes (Fig. 4). Dry conditions in the TX-2019 began in al than the other hybrids. The high biomass sorghum hybrids
Bioenerg. Res.

with lower tissue nutrient concentrations usually have lower observed for tissue K concentrations in IL in this study. The
nutrient removal, and have more sustainable benefits for the limited response of P and K concentrations to N input was
cropping system [45, 51]. presumably due to (1) the substantial environmental impact on
The composition analysis (Fig. 4 and Table 9) indicated tissue nutrient concentrations, and (2) a sufficiency supply of
that the TX17600 hybrid had a juicier stalk and higher sucrose P and K nutrients in the soil for plant uptake [45, 59]. For
concentrations than other hybrids, and its environmentally nutrient removal, plant N removal increased with more N
insensitive and MPS characteristics are likely favorable for input because of the increases in both biomass yield and N
stable and high yielding production in regions with an extend- concentration [34, 44, 56]. Compared to TX, the increased N
ed harvest season (e.g., subtropical or tropical areas) [3]. removal rate in IL was due to higher yield and tissue concen-
Understanding the feedstock’s composition is also essential tration responses to the N fertilizer. The N removal rate in TX
for understanding the quality and quantity of biofuel products, (~0.2 kg N ha−1 uptake per 1.0 kg N input) was similar in
where different quality attributes can be used as indices for several published studies [29, 34, 56]. In TX, the increased
different conversion technologies [15, 52]. Among hybrids, K removal with increasing N input was a result of the DM
the TX17800 hybrid consistently resulted in the highest con- biomass accumulation [45]. Several studies showed that most
centrations of the energy-rich components. In temperate re- of the feedstock components were insensitive to different N
gions such as IL, the TX17800 hybrids can be considered a input [20, 57, 60, 61]. With increasing N input, the improved
good candidate for lignocellulosic energy crops using either lignin content potentially increased the yields of biofuel prod-
bio-processes or thermo-processes because of the excellent ucts, such as crude oil; however, the increased ash content
yield potential along with high feedstock quality [15]. In trop- might lead to an adverse impact in the conversion effective-
ical or subtropical regions of TX, the TX17600 produced ness [15, 62]. The reduced sucrose concentrations occurring at
steady yield with high carbohydrate-rich compounds a high N rate were also observed by Almodares et al. [26].
(Table 9). Although the two VPS hybrids in IL did not show Nitrogen use efficiency can be evaluated based on
distinct differences in terms of yield potential, the TX08001 the biomass DM accumulation per unit of N in the plant
hybrid had a higher feedstock quality than TX17500 by in- tissues (PNUE), and responses of biomass yield and
creasing the energy potential (higher BIC and TIC concentra- crop nutrient removal to the applied N fertilizer (NIE
tions) and lowering ash concentrations (Table 9). The first- and NRE) for optimizing N management [31, 34, 56,
generation TX08001 energy sorghum hybrids have been con- 63]. Variations in the growing environment can be the
sistently reported for both high yield and quality potentials in main factor influencing NUE as well as the associated
several studies [7, 12, 31]. The acetic acid, derived from the management strategies for the energy sorghum cultiva-
acetyl functional group, can be produced during the hydroly- tion [34, 56]. In IL, the increase in PNUE in 2018 was
sis processes (e.g., the dilute acid or enzymatic hydrolysis attributed to the improved DM yield corresponding to
pretreatment), and the increased acetic acid likely inhibits the lowered biomass N concentrations, which was likely
the fermentation effectiveness (biological inhibitor) and cor- due to a favorably growing condition, such as a suffi-
rodes the processing pipelines [52, 53]. In both locations, no cient water supply [34]. Likewise, the severe water
differences in acetyl concentration were observed among hy- stress during the early growing season delayed crop de-
brids (Table 9). velopment and limited the yield response to N supply,
resulting in the lowest NIE in the TX-2018 region. For
Nitrogen Effects and Use Efficiency both locations, favorable weather conditions also led to
higher NRE in 2019 than in 2018, resulting in a 51%
Nitrogen, a primary nutrient for synthesizing amino acids, NRE in 2019. This study showed a similar range of
nucleic acids, or other essential organic compounds, facilitates PNUE (78.1-256.4 kg DM kg−1) to that observed by
crop growth. Weak interactions between N rate and other fac- Grennel et al. [64] and Maw et al. [34], and the
tors for biomass yield indicated that the four hybrids were PNUE likely improved with the lower N input [34,
insensitive to N treatments and possibly had similar N absorb- 65, 66]. Although the N effect on NIE and NRE was
ing abilities even under different environments. Increased N not significant in this study, the better NIE and NRE
supply generally improved biomass yield for all hybrids in the likely occurred at the lower N rate [34, 56, 65–67].
two locations and the recommended N rate for significant Many studies have shown that biomass sorghum has a
yield improvement was less than 120 kg N ha−1, which was low N fertilizer requirement (mostly < 120 kg N ha−1)
also shown in other studies [4, 20, 26, 54–57]. An increase in for optimizing biomass yield with desirable feedstock
biomass N concentrations with increasing N supply shown in qualities for biofuel production [29, 56, 57, 67–69].
this study was often reported in other studies [34, 44, 56, 58]. Moreover, the biomass yield, PNUE, and NIE can be
The improved biomass yield can dilute P and K concentra- further improved by extending the growing season [11,
tions in plant tissues [44, 45, 56]; however, this was only 12, 31]. Even though a low N input was favorable in
 Bioenerg. Res.

this study, the adequate N supply could facilitate PS Declarations

energy sorghum to achieve its maximum yield potential
and nutrient use efficiency by providing the best grow- Disclaimer Any opinions, findings, and conclusions or recommenda-
tions expressed in this publication are those of the author(s) and do not
ing conditions (e.g., irrigation, early planting for extend-
necessarily reflect the views of the US Department of Energy.
ing season length) for the optimum biomass production
and N accumulation [4, 7, 12, 35].
References
1. United States Department of Energy (USDOE) (2016) 2016 billion-
Conclusion ton report: advancing domestic resources for a thriving
bioeconomy, volume 1: economic availability of feedstocks.
This study found cultivation location, year, and hybrid Langholtz MH, Stokes BJ, Eaton LM (Leads), ORNL/TM-2016/
160. Economic Availability of Feedstocks, Oak Ridge National
to significantly influence biomass yield of photoperiod-
Laboratory, Oak Ridge, TN
sensitive sorghum. The 2-year-averaged yields of the 2. Zegada-Lizarazu W, Monti A (2012) Are we ready to cultivate
two VPS sorghum hybrids (TX08001 and TX17500) sweet sorghum as a bioenergy feedstock? A review on field man-
grown in the IL site were 10% and 22% higher, respec- agement practices. Biomass Bioenergy 40:1–12. https://doi.org/10.
tively than in the TX site. Compared to TX, the more 1016/j.biombioe.2012.01.048
3. Rooney WL (2014) Sorghum. In: Karlen DL (ed) Cellulosic energy
consistent weather patterns in IL resulted in more con- cropping systems, 1st edn. Wiley, New York, pp 109–129
sistent performance in biomass yields among hybrids, 4. Rooney WL, Blumenthal J, Bean B, Mullet JE (2007) Designing
which is favorable for a stable feedstock production sorghum as a dedicated bioenergy feedstock. Biofuels Bioprod
and nitrogen (N) management. The TX17600 performed Biorefin 1:147–157. https://doi.org/10.1002/bbb.15
5. Rooney WL, Aydin S (1999) Genetic control of a photoperiod-
consistently in biomass production (with the mean an- sensitive response in Sorghum bicolor (L.) Moench. Crop Sci 39:
nual yield of 21 Mg DM ha−1) and in feedstock com- 397–400. https://doi.org/10.2135/cropsci1999.
position over the four growing environments, displaying 0011183X0039000200016x
resilience to adverse conditions. The four sorghum hy- 6. Miller FR, Quinby JR, Cruzado HJ (1968) Expression of known
maturity genes of sorghum in temperature and tropical environ-
brids showed similar yield responses to N rate, and the
ments1. Crop Sci 8:675–677. https://doi.org/10.2135/cropsci1968.
lower N input (< 112 kg N ha−1) tended to have a 0011183X000800060010x
higher N use efficiency in both locations. Each hybrid 7. Olson SN, Ritter K, Rooney WL, Kemanian A, McCarl BA, Zhang
showed its strength to serve as a bioenergy resource Y, Hall S, Packer D, Mullet J (2012) High biomass yield energy
based on the end-use conversion process (i.e., thermo- sorghum: developing a genetic model for C4 grass bioenergy crops.
Biofuels Bioprod Biorefin 6:640–655. https://doi.org/10.1002/bbb.
chemical or biochemical). The TX17800 hybrid consis- 1357
tently resulted in the highest concentrations of the 8. Mullet J, Morishige D, McCormick R, Truong S, Hilley J,
energy-rich components mainly contributed by the glu- McKinley B, Anderson R, Olson SN, Rooney W (2014) Energy
can, xylan, and lignin. The TX17600 hybrid showed the Sorghum—a genetic model for the design of C4 grass bioenergy
crops. J Exp Bot 65:3479–3489. https://doi.org/10.1093/jxb/
lowest ash content among hybrids. The location and eru229
growing season environment also influenced chemical 9. Turhollow AF, Webb EG, Downing ME (2010) Review of sor-
composition. Drought stress increased non-structural ghum production practices: applications for bioenergy. Oak Ridge
components in the biomass. This increase in non- National Laboratory, Oak Ridge, Tennessee 37831–6283.
Managed by UT-BATTELLE, LLC for the U.S. Department of
structural components possibly decreases conversion ef-
Energy under contract DE-AC05-00OR22725
ficiencies for numerous biomass conversion techniques. 10. Maughan M, Voigt T, Parrish A, Bollero G, Rooney WL, Lee DK
Testing hybrids in the regions of conversion facilities (2012) Forage and energy sorghum responses to nitrogen fertiliza-
would be beneficial to the biofuel industry as results tion in central and southern Illinois. Agron J 104:1032–1040.
https://doi.org/10.2134/agronj2011.0408
varied with the hybrids evaluated in this trial.
11. Meki MN, Ogoshi RM, Kiniry JR, Crow SE, Youkhana AH,
Bioenergy sorghum is a viable annual biomass feed- Nakahata MH, Littlejohn K (2017) Performance evaluation of bio-
stock for the Midwest corn belt and cotton belt. mass sorghum in Hawaii and Texas. Ind Crop Prod 103:257–266.
Drought tolerance, low N requirements, and biomass https://doi.org/10.1016/j.indcrop.2017.04.014
yields of 20 Mg DM ha−1 are all benefits of the adop- 12. McKinley BA, Olson SN, Ritter KB et al (2018) Variation in ener-
gy sorghum hybrid TX08001 biomass composition and lignin
tion of bioenergy sorghum. chemistry during development under irrigated and non-irrigated
field conditions. PLoS One 13:e0195863. https://doi.org/10.1371/
journal.pone.0195863
Funding This work was funded by the DOE Center for Advanced 13. Hoffmann L, Rooney WL (2014) Accumulation of biomass and
Bioenergy and Bioproducts Innovation (US Department of Energy, compositional change over the growth season for six photoperiod
Office of Science, Office of Biological and Environmental Research un- sorghum lines. BioEnergy Res 7:811–815. https://doi.org/10.1007/
der Award Number DE-SC0018420). s12155-013-9405-5
Bioenerg. Res.

14. Bennett AS, Anex RP (2009) Production, transportation and milling 29. Tamang PL, Bronson KF, Malapati A, Schwartz R, Johnson J,
costs of sweet sorghum as a feedstock for centralized bioethanol Moore-Kucera J (2011) Nitrogen requirements for ethanol produc-
production in the upper Midwest. Bioresour Technol 100:1595– tion from sweet and photoperiod sensitive sorghums in the southern
1607. https://doi.org/10.1016/j.biortech.2008.09.023 high plains. Agron J 103:431–440. https://doi.org/10.2134/
15. Li C, Aston JE, Lacey JA, Thompson VS, Thompson DN (2016) agronj2010.0288
Impact of feedstock quality and variation on biochemical and ther- 30. Fazio S, Monti A (2011) Life cycle assessment of different
mochemical conversion. Renew Sust Energ Rev 65:525–536. bioenergy production systems including perennial and annual
https://doi.org/10.1016/j.rser.2016.06.063 crops. Biomass Bioenergy 35:4868–4878. https://doi.org/10.1016/
16. Hoffmann L (2012) Introduction and selection of photoperiod sen- j.biombioe.2011.10.014
sitive sorghum genotypes for agronomic fitness and biomass com- 31. Olson SN, Ritter K, Medley J et al (2013) Energy sorghum hybrids:
position. Dissertation, Texas A&M University functional dynamics of high nitrogen use efficiency. Biomass
17. Bridgwater AV (2012) Review of fast pyrolysis of biomass and Bioenergy 56:307–316. https://doi.org/10.1016/j.biombioe.2013.
product upgrading. Biomass Bioenergy 38:68–94. https://doi.org/ 04.028
10.1016/j.biombioe.2011.01.048 32. Wolfrum E, Payne C, Stefaniak T et al (2013) Multivariate calibra-
18. Kenney KL, Smith WA, Gresham GL, Westover TL (2013) tion models for sorghum composition using near-infrared spectros-
Understanding biomass feedstock variability. Biofuels 4:111–127. copy. Technical Report NREL/TP-510056838. Golden, CO:
https://doi.org/10.4155/bfs.12.83 National Renewable Energy Laboratory (NREL)
19. McBee GG, Creelman RA, Miller FR (1988) Ethanol yield and 33. Maranville JW, Clark RB, Ross WM (1980) Nitrogen efficiency in
energy potential of stems from a spectrum of sorghum biomass grain sorghum. J Plant Nutr 2:577–589. https://doi.org/10.1080/
types. Biomass 17:203–211. https://doi.org/10.1016/0144- 01904168009362800
4565(88)90114-X 34. Maw MJW, Houx JH III, Fritschi FB (2017) Nitrogen use efficien-
20. Amaducci S, Monti A, Venturi G (2004) Non-structural carbohy- cy and yield response of high biomass sorghum in the lower
drates and fibre components in sweet and fibre sorghum as affected Midwest. Agron J 109:115–121. https://doi.org/10.2134/
by low and normal input techniques. Ind Crop Prod 20:111–118. agronj2016.01.0044
https://doi.org/10.1016/j.indcrop.2003.12.016 35. Sadras VO, Lemaire G (2014) Quantifying crop nitrogen status for
21. Gill JR, Burks PS, Staggenborg SA, Odvody GN, Heiniger RW, comparisons of agronomic practices and genotypes. Field Crop Res
Macoon B, Moore KJ, Barrett M, Rooney WL (2014) Yield results 164:54–64. https://doi.org/10.1016/j.fcr.2014.05.006
and stability analysis from the sorghum regional biomass feedstock 36. SAS Institute (2007) SAS/STAT 9.2 Users’s guide. SAS Inst, Cary,
trial. BioEnergy Res 7:1026–1034. https://doi.org/10.1007/s12155- NC
014-9445-5 37. Zhao YL, Dolat A, Steinberger Y, Wang X, Osman A, Xie GH
22. Lee DK, Aberle E, Anderson EK, Anderson W, Baldwin BS, (2009) Biomass yield and changes in chemical composition of
Baltensperger D, Barrett M, Blumenthal J, Bonos S, Bouton J, sweet sorghum cultivars grown for biofuel. Field Crop Res 111:
Bransby DI, Brummer C, Burks PS, Chen C, Daly C, Egenolf J, 55–64. https://doi.org/10.1016/j.fcr.2008.10.006
Farris RL, Fike JH, Gaussoin R, Gill JR, Gravois K, Halbleib MD, 38. Wolabu TW, Tadege M (2016) Photoperiod response and floral
Hale A, Hanna W, Harmoney K, Heaton EA, Heiniger RW, transition in sorghum. Plant Signal Behav 11:e1261232. https://
Hoffman L, Hong CO, Kakani G, Kallenbach R, Macoon B, doi.org/10.1080/15592324.2016.1261232
Medley JC, Missaoui A, Mitchell R, Moore KJ, Morrison JI, 39. Hatfield JL, Prueger JH (2004) Impacts of changing precipitation
Odvody GN, Richwine JD, Ogoshi R, Parrish JR, Quinn L, patterns on water quality. J Soil Water Conserv 59:51–58
Richard E, Rooney WL, Rushing JB, Schnell R, Sousek M,
40. Assefa Y, Staggenborg SA, Prasad VPV (2010) Grain sorghum
Staggenborg SA, Tew T, Uehara G, Viands DR, Voigt T,
water requirement and responses to drought stress: a review. Crop
Williams D, Williams L, Wilson LT, Wycislo A, Yang Y, Owens
Management 9:1–11. https://doi.org/10.1094/cm-2010-1109-01-rv
V (2018) Biomass production of herbaceous energy crops in the
United States: field trial results and yield potential maps from the 41. Song Y, Jain AK, Landuyt W, Kheshgi HS, Khanna M (2015)
multiyear regional feedstock partnership. GCB Bioenergy 10:698– Estimates of biomass yield for perennial bioenergy grasses in the
716. https://doi.org/10.1111/gcbb.12493 USA. BioEnergy Res 8:688–715. https://doi.org/10.1007/s12155-
014-9546-1
23. Rivera-Burgos LA, Volenec JJ, Ejeta G (2019) Biomass and
bioenergy potential of brown midrib sweet sorghum germplasm. 42. Barbanti L, Grandi S, Vecchi A, Venturi G (2006) Sweet and fibre
Front Plant Sci 10:1142. https://doi.org/10.3389/fpls.2019.01142 sorghum (Sorghum bicolor (L.) Moench), energy crops in the frame
of environmental protection from excessive nitrogen loads. Eur J
24. Rocateli AC, Raper RL, Balkcom KS, Arriaga FJ, Bransby DI
Agron 25:30–39. https://doi.org/10.1016/j.eja.2006.03.001
(2012) Biomass sorghum production and components under differ-
ent irrigation/tillage systems for the southeastern U.S. Ind Crop 43. Propheter J, Staggenborg S (2010) Performance of annual and pe-
Prod 36:589–598. https://doi.org/10.1016/j.indcrop.2011.11.007 rennial biofuel crops: nutrient removal during the first two years.
Agron J 102:798–805. https://doi.org/10.2134/agronj2009.0462
25. Packer D (2011) High-biomass sorghums for biomass fuel produc-
tion. Dissertation, Texas A&M University 44. Heitman AJ, Castillo MS, Smyth TJ, Crozier CR, Wang Z,
26. Almodares A, Hadi MR (2009) The effects of nitrogen fertilizer on Heiniger RW, Gehl RJ (2017) Nitrogen fertilization effects on yield
chemical compositions in corn and sweet sorghum. American- and nutrient removal of biomass and sweet sorghum. Agron J 109:
Eurasian J Agric & Environ Sci 6:441–446 1352–1358. https://doi.org/10.2134/agronj2016.12.0710
27. Rathke GW, Diepenbrock W (2006) Energy balance of winter oil- 45. Maw JW, Houx JH III, Fritschi FB (2020) Nitrogen fertilization of
seed rape (Brassica napus L.) cropping as related to nitrogen supply high biomass sorghum affects macro- and micronutrient accumula-
and preceding crop. Eur J Agron 24:35–44. https://doi.org/10.1016/ tion and tissue concentrations. Ind Crop Prod 156:112819. https://
j.eja.2005.04.003 doi.org/10.1016/j.indcrop.2020.112819
28. Rathke GW, Wienhold BJ, Wilhelm WW, Diepenbrock W (2007) 46. Fernandes G, Braga TG, Fisher J et al (2014) Evaluation of poten-
Tillage and rotation effect on corn–soybean energy balances in tial ethanol production and nutrients for four varieties of sweet
eastern Nebraska. Soil Tillage Res 97:60–70. https://doi.org/10. sorghum during maturation. Renew Energy 71:518–524. https://
1016/j.still.2007.08.008 doi.org/10.1016/j.renene.2014.05.033
 Bioenerg. Res.

47. Monti A, Di Virgilio N, Venturi G (2008) Mineral composition and 58. Beyaert RP and Roy RC (2005) Influence of nitrogen fertilization
ash content of six major energy crops. Biomass Bioenergy 32:216– on multi-cut forage sorghum–sudangrass yield and nitrogen use.
223. https://doi.org/10.1016/j.biombioe.2007.09.012 Agron J 97:1493–1501. https://doi.org/10.2134/agronj2005.0079
48. Emerson R, Hoover A, Ray A, Lacey J, Cortez M, Payne C, Karlen 59. Soileau JM and Bradford BN (1985) Biomass and sugar yield re-
D, Birrell S, Laird D, Kallenbach R, Egenolf J, Sousek M, Voigt T sponse of sweet sorghum to lime and fertilizer. Agron J 77:471–
(2014) Drought effects on composition and yield for corn stover, 475. https://doi.org/10.2134/agronj1985.
mixed grasses, and Miscanthus as bioenergy feedstocks. Biofuels 5: 00021962007700030025x
17–291. https://doi.org/10.1080/17597269.2014.913904 60. Marsalis MA, Angadi SV, Contreras-Govea FE (2010) Dry matter
49. Ong RG, Higbee A, Bottoms S, Dickinson Q, Xie D, Smith SA, yield and nutritive value of corn, forage sorghum, and BMR forage
Serate J, Pohlmann E, Jones AD, Coon JJ, Sato TK, Sanford GR, sorghum at different plant populations and nitrogen rates. Field
Eilert D, Oates LG, Piotrowski JS, Bates DM, Cavalier D, Zhang Y Crop Res 116:52–57. https://doi.org/10.1016/j.fcr.2009.11.009
(2016) Inhibition of microbial biofuel production in 61. Anfinrud R, Cihacek L, Johnson BL, Ji Y, Berti MT (2013)
droughtstressed switchgrass hydrolysate. Biotechnol Biofuels 9: Sorghum and kenaf biomass yield and quality response to nitrogen
237–250. https://doi.org/10.1186/s13068-016-0657-0 fertilization in the Northern Great Plains of the USA. Ind Crop Prod
50. Hoover A, Emerson R, Ray A, Stevens D, Morgan S, Cortez M, 50:159–165. https://doi.org/10.1016/j.indcrop.2013.07.022
Kallenbach R, Sousek M, Farris R, Daubaras D (2018) Impact of 62. Mahmud K, Ahmad I, Ayub M (2003) Effect of nitrogen and phos-
drought on chemical composition and sugar yields from dilute-acid phorus on the fodder yield and quality of two sorghum cultivars
pretreatment and enzymatic hydrolysis of miscanthus, a tall fescue (Sorghum bicolor L.). Int J Agric Biol 5:61–63
mixture, and switchgrass. Front Energy Res 6:54. https://doi.org/
63. Zweifel TR, Maranville JW, Ross WM et al (1987) Nitrogen fertil-
10.3389/fenrg.2018.00054
ity and irrigation influence on grain sorghum nitrogen efficiency.
51. Singh MP, Erickson JE, Sollenberger LE, Woodard KR,
Agron J 79:419–422. https://doi.org/10.2134/agronj1987.
Vendramini JMB, Fedenko JR (2012) Mineral composition and
00021962007900030001x
biomass partitioning of sweet sorghum grown for bioenergy in
the southeastern USA. Biomass Bioenergy 47:1–8. https://doi. 64. Grennel JL (2014) Yield and carbon exchange of sorghum grown as
org/10.1016/j.biombioe.2012.10.022 advanced biofuel feedstock on abandoned agricultural land in
52. Jönsson LJ, Alriksson B, Nilvebrant NO (2013) Bioconversion of southeastern Ohio. Thesis. Ohio Univ, Athens
lignocellulose: inhibitors and detoxification. Biotechnol Biofuels 6: 65. Maranville JW, Pandey RK, and Sirifi S (2002) Comparison of
16. https://doi.org/10.1186/1754-6834-6-16 nitrogen use efficiency of a newly developed sorghum hybrid and
53. Foust TD, Aden A, Dutta A, Phillips S (2009) An economic and two improved cultivars in the Sahel of West Africa. Commun Soil
environmental comparison of a biochemical and a thermochemical Sci Plant Anal 33:1519–1536. https://doi.org/10.1081/CSS-
lignocellulosic ethanol conversion processes. Cellulose 16:547– 120004298
565. https://doi.org/10.1007/s10570-009-9317-x 66. Holman JD, Obour AK, and Mengel DB (2019) Nitrogen applica-
54. Zhao D, Reddy KR, Kakani VG, Reddy VR (2005) Nitrogen defi- tion effects on forage sorghum production and nitrate concentration.
ciency effects on plant growth, leaf photosynthesis, and J Plant Nutr 42:2794–2804. https://doi.org/10.1080/01904167.
hyperspectral reflectance properties of sorghum. Eur J Agron 22: 2019.1659321
391–403. https://doi.org/10.1016/j.eja.2004.06.005 67. Maw MJW, Houx JH, Fritschi FB (2019) Nitrogen content and use
55. Olugbemi O, Abiola Ababyomi Y (2016) Effects of nitrogen appli- efficiency of sweet sorghum grown in the lower Midwest. Agron J
cation on growth and ethanol yield of sweet sorghum [Sorghum 111:2920–2928. https://doi.org/10.2134/agronj2018.08.0489
bicolor (L.) Moench] varieties. Adv Agric 2016:8329754. https:// 68. Wiedenfeld RP (1984) Nutrient requirements and use efficiency by
doi.org/10.1155/2016/8329754 sweet sorghum. Energy Agr 3:49–59. https://doi.org/10.1016/
56. Ameen A, Yang X, Chen F, Tang C, Du F, Fahad S, Xie GH (2017) 0167-5826(84)90004-4
Biomass yield and nutrient uptake of energy sorghum in response to 69. Adams CB, Erickson JE, Singh MP (2015) Investigation and syn-
nitrogen fertilizer rate on marginal land in a semi-arid region. thesis of sweet sorghum crop responses to nitrogen and potassium
BioEnergy Res 10:363–376. https://doi.org/10.1007/s12155-016- fertilization. Field Crop Res 178:1–7. https://doi.org/10.1016/j.fcr.
9804-5 2015.03.014
57. Tang C, Yang X, Chen X, Ameen A, Xie G (2018) Sorghum bio-
mass and quality and soil nitrogen balance response to nitrogen rate
Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
on semiarid marginal land. Field Crop Res 215:12–22. https://doi.
tional claims in published maps and institutional affiliations.
org/10.1016/j.fcr.2017.09.031